Method for detecting tool state of numerical control machine tool, electronic device and storage medium

Abstract

The embodiment of the application provides a kind of detection method of tool state of numerical control machine tool, electronic equipment and storage medium, related to machine tool processing technical field, can be in the case where not affecting material processing, realize the real-time detection of tool state, to facilitate timely discovery sudden broken tool event, ensure the yield of the material processed.Therein, the detection method of tool state of numerical control machine tool includes: obtaining the previous vibration time series data and the current vibration time series data generated in the machining process of the current tool of numerical control machine tool;According to previous vibration time series data and current vibration time series data, respectively, obtain the vibration feature matrix corresponding to previous vibration time series data and current vibration time series data;According to the vibration feature matrix corresponding to previous vibration time series data and current vibration time series data, obtain the feature difference degree between the previous machining and the current machining process of the current tool;Based on the comparison result of feature difference degree and current difference degree threshold, determine the current state of the current tool.

Description

Technical Field

[0001] This application relates to the field of machine tool processing technology, and in particular to a method, electronic device and storage medium for detecting the tool status of a CNC machine tool. Background Technology

[0002] CNC machine tools process materials using their own installed cutting tools, and the condition of these tools determines the quality and precision of the processed materials. Therefore, monitoring the condition of the cutting tools is essential to ensuring the quality and precision of the processed materials.

[0003] Currently, tool setting devices are often embedded in CNC machine tools. After the tool has processed a certain amount of material, processing is paused, and the tool status is then checked by the tool setting device. This method has two main problems: First, each check by the tool setting device is time-consuming, affecting material processing efficiency; second, the tool setting device performs tool breakage checks at regular intervals, which may result in the failure to detect tool breakage in a timely manner, thus affecting the quality and precision of the processed materials. Summary of the Invention

[0004] This application provides a method, electronic device, and storage medium for detecting the tool status of a CNC machine tool. It can realize real-time detection of the tool status without affecting material processing, so as to promptly detect sudden tool breakage events and ensure the yield of processed materials.

[0005] In a first aspect, embodiments of this application provide a method for detecting the tool condition of a CNC machine tool, the method comprising:

[0006] Acquire the previous vibration timing data and the current vibration timing data generated by the current tool during the machining process of the CNC machine tool;

[0007] Based on the previous vibration time series data and the current vibration time series data, vibration feature matrices corresponding to the previous vibration time series data and the current vibration time series data are obtained respectively; wherein, the vibration feature matrix includes vibration time domain features and vibration frequency domain features;

[0008] Based on the vibration feature matrix corresponding to the previous vibration time sequence data and the current vibration time sequence data, the current feature difference degree of the current tool between the previous processing and the current processing is obtained.

[0009] Based on the comparison results between the current feature difference degree and the current difference degree threshold, the current state of the current tool is determined.

[0010] In this embodiment, by acquiring the previous vibration timing data and the current vibration timing data generated by a certain tool in a CNC machine tool during the two most recent machining processes, and then acquiring the time-frequency domain features corresponding to the previous and current vibration timing data respectively, the feature difference degree between the current and previous vibration timing data can be determined relatively accurately using the time-frequency domain features corresponding to the previous and current vibration timing data. That is, the feature difference degree reflects the degree of change of the current vibration timing data compared to the previous vibration timing data. The smaller the feature difference degree, the smaller the degree of change of the current vibration timing data compared to the previous vibration timing data; conversely, the larger the feature difference degree, the greater the degree of change of the current vibration timing data compared to the previous vibration timing data. Finally, by comparing the above feature difference degree with the current difference degree threshold, the current tool state corresponding to the generation of the current vibration data is determined. This method allows for real-time detection of the tool status without affecting CNC machine tool processing, thereby promptly identifying sudden tool breakage events and ensuring the quality and precision of the processed materials.

[0011] Optionally, the step of obtaining the vibration feature matrices corresponding to the previous vibration time series data and the current vibration time series data based on the previous vibration time series data and the current vibration time series data respectively includes:

[0012] The previous vibration time series data and the current vibration time series data are filtered based on the first target frequency band;

[0013] Based on the filtered previous vibration time series data and the current vibration time series data, the vibration time domain features and vibration frequency domain features corresponding to the previous vibration time series data and the current vibration time series data are extracted respectively.

[0014] In this embodiment, the first target frequency band can be considered as the working frequency band of the tool during the current processing stage. Filtering the previous vibration timing data and the current vibration timing data generated in the two most recent processing stages using the first target frequency band helps to eliminate other frequency bands unrelated to the processing stage, making the extracted time-frequency domain features for the processing stage more accurate based on the previous vibration timing data and the current vibration timing data.

[0015] Optionally, based on the filtered previous vibration time-series data and the current vibration time-series data, the vibration time-domain features and vibration frequency-domain features corresponding to the previous vibration time-series data and the current vibration time-series data are extracted respectively, including:

[0016] The fundamental frequency for time-frequency domain conversion is determined based on the current set rotational speed of the tool.

[0017] Based on the current number of cutting edges of the tool and the fundamental frequency, determine the maximum harmonic of the time-frequency domain conversion;

[0018] Based on the fundamental frequency and the maximum harmonic, at least two second target frequency bands for time-frequency domain conversion are determined;

[0019] The filtered previous vibration time series data and the current vibration time series data are respectively converted into time and frequency domains to obtain the vibration frequency domain features corresponding to each second target frequency band in the previous vibration time series data and the current vibration time series data.

[0020] In this embodiment, based on the current tool's set rotation speed and number of cutting edges, multiple second target frequency bands containing effective signals of tool vibration can be determined. Then, the frequency domain features of each second target frequency band are obtained, thereby improving the accuracy of the extracted frequency domain features.

[0021] Optionally, determining the current state of the tool based on the comparison result between the current feature difference degree and the current difference degree threshold includes:

[0022] Determine whether the current feature difference is greater than the current difference threshold;

[0023] If the difference is greater than the current difference threshold, then the current tool is determined to be in an abnormal state.

[0024] In this embodiment, if the difference between the current vibration timing data and the previous vibration timing data is large, it indicates that the current vibration timing data has changed significantly compared to the previous vibration timing data. If the tool state corresponding to the previous vibration timing data was normal, then it can be inferred that the tool state at the time the current vibration timing data was generated was abnormal. If the tool state corresponding to the previous vibration timing data was abnormal, and if the difference in the current feature of the current tool is greater than the current difference threshold, then the tool state at the time the current vibration timing data was generated is still abnormal.

[0025] Optionally, after determining whether the current feature difference is greater than the current difference threshold, the method further includes:

[0026] If the difference is not greater than the current difference threshold, then the current tool is determined to be in a normal state.

[0027] In this embodiment, if the difference between the current vibration timing data and the previous vibration timing data is small, it indicates that the change in the current vibration timing data compared to the previous vibration timing data is small. If the tool state corresponding to the previous vibration timing data is normal, then it can be inferred that the tool state when the current vibration timing data was generated was normal. If the tool state corresponding to the previous vibration timing data is abnormal, and if the difference in the current feature of the current tool is not greater than the current difference threshold, then the tool state when the current vibration timing data was generated is still normal.

[0028] Optionally, before determining the current state of the tool based on the comparison result between the current feature difference degree and the current difference degree threshold, the method further includes:

[0029] Acquire multiple vibration timing data of the processing procedure with a previously set number of cycles;

[0030] Obtain the multiple vibration feature matrix corresponding to the multiple vibration time series data;

[0031] Based on the vibration feature matrix corresponding to the multiple vibration time series data, the feature difference degree between two adjacent processing operations in the previous set number of processing operations of the CNC machine tool is obtained;

[0032] The current difference threshold is updated based on the feature difference between two adjacent processing operations during the previously set number of processing operations.

[0033] In this embodiment of the application, due to the differences in the workpiece during the current tool processing, the current difference threshold may not be able to distinguish the current tool state well. Therefore, after each set number of processing operations, the current difference threshold will be updated to improve the accuracy of tool state judgment.

[0034] Optionally, updating the current difference threshold based on the feature difference between two adjacent processes in the previous processing includes:

[0035] Obtain the feature difference between two adjacent processing steps in the previous processing steps, and construct a difference set;

[0036] Based on the mean and variance of the feature differences corresponding to the difference set, and the size parameters of the current tool, obtain the threshold to be updated;

[0037] Determine whether the threshold to be updated is greater than the initial difference threshold;

[0038] If it is greater than the threshold to be updated, then the threshold to be updated will be used as the current difference threshold.

[0039] If it is not greater than, then the initial difference threshold is used as the current difference threshold.

[0040] In this embodiment, for several previous machining processes, the overall similarity mean and overall similarity variance corresponding to multiple feature differences can be determined based on the feature differences between vibration time-series data generated by two adjacent machining processes. Then, an update threshold is determined based on the overall similarity mean, overall similarity variance, and the current tool size parameters. If the update threshold is greater than the initial difference threshold obtained from previous training, the update threshold is used as the new current difference threshold. With a larger current difference threshold, the detection adaptability is modified to better detect the tool state. If the update threshold is not greater than the initial difference threshold obtained from previous training, the initial difference threshold is used as the new current difference threshold, that is, the lower limit of the threshold is used to detect the tool state.

[0041] Optionally, before determining the current state of the tool based on the comparison result between the current feature difference degree and the current difference degree threshold, the method further includes:

[0042] Acquire the historical vibration timing data and tool status data generated by the tool during the machining process of the CNC machine tool.

[0043] Obtain the vibration feature matrix corresponding to the vibration time series data of each vibration;

[0044] Based on the vibration feature matrix corresponding to the vibration time series data, the feature difference degree corresponding to two adjacent processing times in the historical processing of the CNC machine tool is obtained, and a set of feature difference degrees for each time is constructed.

[0045] With the aim of distinguishing between abnormal and normal tool states by the initial difference threshold, the initial difference threshold is iteratively adjusted based on the previous feature difference set and the tool state until the iteration ends, and the initial difference threshold is obtained.

[0046] The initial difference threshold is used as the initial value of the current difference threshold.

[0047] In this embodiment, an initial difference threshold that can effectively distinguish between abnormal and normal tool states can be learned based on the historical vibration time sequence data and tool state data generated by the tool during the CNC machine tool's historical machining process, and this threshold can be used as the initial value of the current difference threshold.

[0048] In a second aspect, embodiments of this application provide an electronic device, the electronic device including a memory for storing a computer program and a processor for executing the computer program, wherein when the computer program is executed by the processor, the electronic device is triggered to perform the steps of the method as described in any embodiment of the first aspect.

[0049] Thirdly, embodiments of this application provide a computer-readable storage medium for storing a computer program that, when a processor is run, causes the processor to perform the steps of the method as described in any embodiment of the first aspect.

[0050] It should be understood that the second to fourth aspects of the embodiments of this application are consistent with the technical solutions of the first aspect of the embodiments of this application, and the beneficial effects achieved by each aspect and the corresponding feasible implementation are similar, and will not be described again. Attached Figure Description

[0051] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this specification. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0052] Figure 1 A flowchart illustrating a method for detecting the tool status of a CNC machine tool, provided in an embodiment of this application;

[0053] Figure 2 A flowchart illustrating a method for extracting vibration feature matrices provided in an embodiment of this application;

[0054] Figure 3 A flowchart illustrating a method for extracting frequency domain features according to an embodiment of this application;

[0055] Figure 4 A flowchart illustrating a method for determining the current tool state of a cutting tool, provided in an embodiment of this application;

[0056] Figure 5 A flowchart illustrating a method for determining a current difference threshold provided in an embodiment of this application;

[0057] Figure 6 A flowchart illustrating a method for updating the current difference threshold provided in an embodiment of this application;

[0058] Figure 7 A flowchart illustrating the method for updating the current difference threshold provided in an embodiment of this application;

[0059] Figure 8 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0060] To better understand the technical solutions in this specification, the embodiments of this application will be described in detail below with reference to the accompanying drawings.

[0061] It should be understood that the described embodiments are merely some, not all, of the embodiments in this specification. All other embodiments obtained by those skilled in the art based on the embodiments in this specification without inventive effort are within the scope of protection of this specification.

[0062] The terminology used in the embodiments of this application is for the purpose of describing particular embodiments only and is not intended to be limiting of this specification. The singular forms “a,” “the,” and “the” used in the embodiments of this application and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise.

[0063] Currently, the common practice is to embed a tool setting device into the CNC machine tool. After processing a certain amount of material, the machine is paused, and the tool setting device is used to check the tool's condition. This method has two main problems: First, using a tool setting device requires pausing processing, and each check is time-consuming, thus affecting material processing efficiency. Second, if a tool breaks suddenly at regular intervals, the breakage may not be detected in time, affecting the quality and precision of the processed material and increasing the defect rate.

[0064] In view of this, this application provides a method for detecting the tool status of a CNC machine tool. In this method, the vibration time sequence data of the tool in the CNC machine tool during the last two machining processes are acquired in real time. Then, the feature difference degree between the time-frequency domain features of the above vibration time sequence data is calculated, and the feature difference degree is compared with the current difference degree threshold to obtain the tool status detection result. That is, the tool status can be detected in real time without affecting the material processing, so as to promptly detect sudden tool breakage events and ensure the yield of the processed materials.

[0065] The technical solutions protected by the embodiments of this application will be described in detail below with reference to the accompanying drawings.

[0066] Please see Figure 1 This is a flowchart illustrating a tool condition detection method for a CNC machine tool provided in an embodiment of this application. This method is applied to electronic devices, such as personal computers (PCs), workstations, servers, etc. The server can be a general-purpose computer, a special-purpose computer, or a server cluster; this application does not impose any particular limitations on this. The flowchart of this method is described as follows:

[0067] Step 101: Obtain the previous vibration timing data and the current vibration timing data generated by the current tool during the machining process of the CNC machine tool.

[0068] In this embodiment, a vibration sensor can be installed on the spindle of a CNC machine tool to collect vibration timing data generated by the tool after the CNC machine tool is started. It should be understood that the vibration sensor collects three-axis vibration timing data, namely, the vibration timing data of the tool on the X-axis, Y-axis, and Z-axis.

[0069] The vibration timing data collected by the vibration sensor, along with the corresponding processing attribute information, is transmitted to the electronic device. For example, processing attribute information may include machine number, material serial number, tool number, program segment, timestamp, machine status (whether it is waiting for material, alarm, or processing), and machine operation task (whether it is inspection or material processing). The electronic device can filter the received vibration timing data based on the above processing attribute information. For example, it can filter out vibration timing data where the program segment is in processing, and then further filter out vibration timing data for a specific machine and tool number to be detected. Furthermore, it can use the material serial number to filter out the previous vibration timing data generated in the last two processing cycles and the current vibration timing data. Of course, other attribute information can also be combined for filtering; this application does not impose any particular limitations on this.

[0070] Step 102: Based on the previous vibration time series data and the current vibration time series data, obtain the vibration feature matrices corresponding to the previous vibration time series data and the current vibration time series data, respectively.

[0071] In this embodiment, the vibration feature matrix includes vibration time-domain features and vibration frequency-domain features. The vibration time-domain features and vibration frequency-domain features are described in detail below.

[0072] Vibration time-domain characteristics include: amplitude characteristic indicators, statistical characteristic indicators, waveform characteristic indicators, and impact characteristic indicators. Amplitude characteristic indicators include: peak value, trough value, and peak-to-peak value; statistical characteristic indicators include: mean, variance, standard deviation, and root mean square value; waveform characteristic indicators include: kurtosis, skewness, and waveform indices; impact characteristic indicators include: peak value index, impulse index, and marginal coefficient. The following sections will provide a detailed introduction to these thirteen vibration time-domain characteristics.

[0073] Vibration time-domain characteristic 1:

[0074] Peak value: The maximum value of the signal, reflecting the maximum amplitude of the tool vibration. The formula for the peak value is shown in (1):

[0075] f td1 =max(s(t))(1)

[0076] Among them, f td1 denoted by , and s(t) represents the vibration amplitude of any axis at the t-th time point.

[0077] Vibration time-domain feature two:

[0078] Valley value: The minimum value of the signal, reflecting the minimum amplitude of tool vibration. The formula for the valley value is shown in (2):

[0079] f td2 =min(s(t))(2)

[0080] Among them, f td2 This indicates the valley value.

[0081] Vibration time-domain feature three:

[0082] Peak-to-peak value: The difference between the peak value and the trough value reflects the amplitude range of the tool vibration.

[0083] The formula for peak-to-peak value is shown in (3):

[0084] f td3 =f td1 -f td2 (3)

[0085] Among them, f td3 This indicates the peak value.

[0086] Vibration time-domain characteristic four:

[0087] Mean: Represents the average amplitude of the signal, reflecting the average level of tool vibration. The formula for the mean is shown in (4):

[0088]

[0089] Among them, f td4 This represents the mean, and T represents the total number of timestamps.

[0090] Vibration time-domain feature five:

[0091] Variance: Measures the degree to which a signal deviates from the mean, reflecting the stability of tool vibration. The formula for calculating variance is shown in (5):

[0092]

[0093] Among them, f td5 Indicates variance.

[0094] Vibration time-domain feature six:

[0095] Standard deviation: The square root of the variance, also an indicator of tool vibration stability. The formula for calculating the standard deviation is shown in (6):

[0096]

[0097] Among them, f td6Indicates the standard deviation.

[0098] Vibration time-domain characteristic seven:

[0099] Root mean square (RMS) value: The root mean square value of the signal, often used to measure the energy of tool vibration. The formula for calculating the RMS value is shown in (7):

[0100]

[0101] Among them, f td7 This represents the root mean square value.

[0102] Vibration time-domain feature eight:

[0103] Kurtosis: The degree of concentration of signal distribution, describing the concentration near the peak of tool vibration. The formula for calculating kurtosis is shown in (8):

[0104]

[0105] Among them, f td8 Indicates raucity.

[0106] Vibration time-domain feature nine:

[0107] Skewness: The asymmetry of signal distribution, describing the degree of deviation of tool vibration. The formula for calculating skewness is shown in (9):

[0108]

[0109] Among them, f td9 Indicates skewness.

[0110] Vibration time-domain characteristics ten:

[0111] Waveform index: The ratio between the root mean square value of a signal and the absolute value of its average value. Waveform index reflects the characteristics of the cutting tool during machining as follows:

[0112] First, the asymmetry and nonlinearity of signal waveforms: Waveform indices can reveal the symmetry and nonlinearity characteristics of signal waveforms. If the signal is a perfectly symmetrical sine wave, its waveform factor is close to 1. However, for actual tool vibration signals, due to the possible asymmetric impacts and irregular fluctuations, the waveform indices will be greater than 1, and the larger the ratio, the more irregular the signal waveform.

[0113] Second, the complexity of tool vibration: Larger waveform values ​​usually indicate that the tool vibration signal contains more complex components, such as impact, noise, or non-periodic components. This may be related to tool wear, uneven load during the cutting process, or the dynamic characteristics of the machine tool.

[0114] Third, signal energy distribution: The waveform factor is also related to the distribution of signal energy. A higher waveform factor may indicate that the signal energy is concentrated at certain specific moments or frequencies, which may affect the cutting efficiency and machining quality of the tool.

[0115] The formula for calculating the waveform index is shown in formula (10):

[0116]

[0117] Among them, f td10 Indicates waveform index.

[0118] Vibration time-domain feature eleven:

[0119] Peak index: Reflects the impact component in tool vibration. A large peak index indicates a significant impact component in tool vibration, which may be related to tool wear, breakage, or instability during machining. The formula for calculating the peak index is shown in (11):

[0120]

[0121] Among them, f td11 This indicates the peak indicator.

[0122] Vibration time-domain characteristics 12:

[0123] Pulse index: A relative measure between the maximum amplitude or intensity of a signal and the overall energy level or intensity. It reflects the relationship between extreme values ​​and the average value of the signal in terms of amplitude or intensity. The pulse index can reflect the following characteristics during tool machining:

[0124] Tool wear condition: As the tool gradually wears down, parameters such as the maximum cutting force and maximum vibration amplitude during the cutting process may change. Simultaneously, the overall vibration or cutting force level may also be affected. Therefore, this ratio can serve as a sensitive indicator for monitoring tool wear condition. An increase in the ratio may indicate that tool wear has led to an increase in the maximum amplitude or intensity of the vibration relative to the overall energy level during the cutting process.

[0125] Stability of the cutting process: The stability of the cutting process is crucial to the machining quality. The root mean square (RMS) value of the signal reflects the overall vibration or cutting force level during the cutting process, while the maximum absolute value of the signal reflects extreme conditions during the cutting process. Therefore, this ratio can reflect the stability of the cutting process. If the ratio fluctuates greatly, it indicates that there are large amplitude or intensity changes during the cutting process, which may indicate that the cutting process is not stable enough.

[0126] Changes in cutting conditions: Variations in cutting conditions such as cutting speed, feed rate, and depth of cut can affect the maximum cutting force, maximum vibration amplitude, and overall vibration or cutting force level during the cutting process. Therefore, by monitoring changes in this ratio, the impact of changes in cutting conditions on the machining process can be detected in a timely manner.

[0127] The formula for calculating the pulse index is shown in (12):

[0128]

[0129] Among them, f td12 This indicates the pulse index.

[0130] Vibration time-domain feature thirteen:

[0131] Marginal coefficient: In the context of tool machining, the marginal coefficient reflects the relationship between the maximum amplitude and the dispersion of the tool vibration signal. A high marginal coefficient indicates a large peak in the tool vibration, which may mean a large impact or instability in the machining process relative to the average level of vibration. This index can be used to assess the wear condition of the tool or abnormal conditions in the machining process because it can reveal the relationship between the extreme values ​​in the vibration signal and the overall vibration level. The formula for calculating the marginal coefficient is shown in (13):

[0132]

[0133] Among them, f td13 This represents the marginal coefficient.

[0134] Vibration frequency domain characteristics include: power spectrum characteristics, statistical characteristics, and waveform characteristics. Power spectrum characteristics include: peak power density and peak frequency; statistical characteristics include: average power spectral density, power spectral density variance, frequency standard deviation, and root mean square frequency; waveform characteristics include: spectral kurtosis, spectral skewness, frequency skewness, frequency kurtosis, centroid frequency, D-factor, E-factor, and G-factor. The following sections provide a detailed introduction to these fourteen vibration frequency domain characteristics.

[0135] Vibration frequency domain characteristic 1:

[0136] Average power spectral density: The average power spectral density over the entire frequency domain, reflecting the overall energy distribution of tool vibration. The formula for calculating the average power spectral density is shown in (14):

[0137]

[0138] Among them, f tfd1Let p(k) represent the average power spectral density, p(k) represent the power spectral density at the k-th frequency point, and K represent the total number of frequency points.

[0139] Vibration frequency domain characteristic two:

[0140] Power spectral density variance: The degree of dispersion of the power spectral density reflects the difference in energy distribution among different frequency components. The formula for calculating the power spectral density variance is shown in (15):

[0141]

[0142] Among them, f tfd2 This represents the variance of the power spectral density.

[0143] Vibration frequency domain characteristic three:

[0144] Spectral skewness: The asymmetry of the power spectral density distribution reflects the degree of skewness in the tool vibration spectrum. The formula for calculating the spectral skewness is shown in (16):

[0145]

[0146] Among them, f tfd3 Indicates spectral skewness.

[0147] Vibration frequency domain characteristic four:

[0148] Spectral kurtosis: The sharpness of the power spectral density distribution, reflecting the concentration of peaks in the tool vibration spectrum. The formula for calculating spectral skewness is shown in (17):

[0149]

[0150] Among them, f tfd4 Indicates spectral kurtosis.

[0151] Vibration frequency domain characteristic five:

[0152] Center of gravity frequency: the geometric center of the power spectrum, reflecting the main concentrated area of ​​tool vibration energy. The formula for calculating the center of gravity frequency is shown in (18):

[0153]

[0154] Among them, f tfd5 Let f(k) represent the centroid frequency, and f(k) represent the frequency of the k-th frequency point.

[0155] Vibration frequency domain characteristic six:

[0156] Frequency standard deviation: The weighted standard deviation of the spectrum describes the width of the signal's spectral distribution. The formula for calculating the frequency standard deviation is shown in (19):

[0157]

[0158] Among them, f tfd6 This indicates the standard deviation of the frequency.

[0159] Vibration frequency domain characteristic seven:

[0160] Root mean square (RMS) frequency: The weighted root mean square value of the spectrum, which describes the energy center of the signal's spectral distribution. The formula for calculating the RMS frequency is shown in (20):

[0161]

[0162] Among them, f tfd7 This represents the root mean square frequency.

[0163] Vibration frequency domain characteristic eight:

[0164] D-factor: A dimensionless number used to describe the shape of the spectrum. It describes the distribution characteristics of the signal spectrum, especially the sharpness of the spectrum and the influence of high-frequency components. The formula for calculating the D-factor is shown in (21):

[0165]

[0166] Among them, f tfd8 This represents the D factor.

[0167] Vibration frequency domain characteristic nine:

[0168] E-factor: A dimensionless number used to describe the shape of the spectrum. It reflects the influence of the energy center and high-frequency components of the spectrum. The formula for calculating the E-factor is shown in (22):

[0169]

[0170] Among them, f tfd9 This represents the E factor.

[0171] Vibration frequency domain characteristics ten:

[0172] G-factor: A dimensionless number used to standardize the spectral width of a signal, making it independent of the centroid frequency. It reflects the relative width or variability of the spectral distribution. The formula for calculating the G-factor is shown in (23):

[0173]

[0174] Among them, f tfd10 This represents the G factor.

[0175] Vibration frequency domain characteristic eleven:

[0176] Frequency skewness: Represents the weighted skewness of the signal spectrum, describing the asymmetry of the spectral distribution. The formula for calculating frequency skewness is shown in (24):

[0177]

[0178] Among them, f tfd11 Indicates frequency skewness.

[0179] Vibration frequency domain characteristic twelve:

[0180] Frequency kurtosis describes the sharpness of a signal's spectrum, i.e., whether the spectral distribution has high peak values. High kurtosis indicates the presence of significant sharp components in the spectrum, which may correspond to impulses or transient events in the signal. The formula for calculating frequency kurtosis is shown in (25):

[0181]

[0182] Among them, f tfd12 Indicates frequency kurtosis.

[0183] Vibration frequency domain characteristic thirteen:

[0184] Peak power spectral density: The maximum value of the power spectral density, reflecting the power of the most significant frequency component. The peak power spectral density is calculated as shown in formula (26):

[0185] f tfd13 =max(p(k))(26)

[0186] Among them, f tfd13 This represents the peak value of the power spectral density.

[0187] Vibration frequency domain characteristic fourteen:

[0188] Peak frequency: The frequency point with the highest power spectral density, indicating the most significant frequency component in tool vibration. The formula for calculating the peak frequency is shown in (27):

[0189] f tfd14 =argmax(p(k))(27)

[0190] Among them, f tfd14 Indicates the peak frequency.

[0191] In some embodiments, considering that the vibration frequency of the cutting tool during the machining process may be mainly concentrated in a specific frequency band, in this embodiment of the application, the vibration feature matrix can be extracted only from the vibration time series data located in the machining frequency band, thereby improving the accuracy of the extracted vibration feature matrix.

[0192] Please see Figure 2This is a flowchart illustrating a method for extracting a vibration feature matrix according to an embodiment of this application. Step 102 can be specifically implemented by executing sub-steps 1021 to 1022:

[0193] Step 1021: Filter the previous vibration time series data and the current vibration time series data based on the first target frequency band.

[0194] Step 1022: Based on the filtered previous vibration time series data and the current vibration time series data, extract the vibration time domain features and vibration frequency domain features corresponding to the previous vibration time series data and the current vibration time series data, respectively.

[0195] In this embodiment, the first target frequency band can be considered as the working frequency band of the tool during the current machining stage. For example, the first target frequency band can be [50Hz, 1950Hz]. Filtering the previous vibration timing data and the current vibration timing data generated in the two most recent machining processes using the first target frequency band helps to eliminate other frequency bands unrelated to the machining stage, making the extracted time-frequency domain features for the machining stage more accurate based on the previous vibration timing data and the current vibration timing data.

[0196] In some embodiments, considering that the rotational speed and number of cutting edges of the tool during operation may also affect the distribution of its effective vibration signal, in this embodiment, frequency domain features can be extracted within the frequency band where the effective vibration signal of the tool is concentrated, thereby improving the accuracy of the extracted frequency domain features.

[0197] Please see Figure 3 This is a flowchart illustrating a method for extracting frequency domain features according to an embodiment of this application. Step 1022 can be specifically implemented by executing sub-steps 10221 to 10224:

[0198] Step 10221: Determine the fundamental frequency for time-frequency domain conversion based on the current tool's set rotational speed.

[0199] In this embodiment of the application, taking the current tool's set rotation speed of 6000 r / min as an example, the fundamental frequency of the time-frequency domain conversion is 6000 / 60 = 100 Hz.

[0200] Step 10222: Determine the maximum harmonic of the time-frequency domain conversion based on the current number of cutting edges and the fundamental frequency of the tool.

[0201] In this embodiment of the application, taking a tool with four cutting edges as an example, if the base frequency is 100Hz, then the maximum harmonic frequency of the time-frequency domain conversion is 100*4=400Hz.

[0202] Step 10223: Determine at least two second target frequency bands for time-frequency domain conversion based on the fundamental frequency and the maximum harmonic.

[0203] In this embodiment of the application, if the base frequency is 100Hz and the maximum harmonic frequency is 400Hz, then the second target frequency bands can be directly determined as [50Hz, 250Hz] and [250Hz, 450Hz].

[0204] Of course, in another possible embodiment, multiple intermediate harmonics can be determined between the fundamental frequency and the maximum harmonic. Then, based on the fundamental frequency, the multiple intermediate harmonics, and the maximum harmonic, at least two second target frequencies for time-frequency domain conversion can be determined. Continuing with the example of a fundamental frequency of 100Hz and a maximum harmonic of 400Hz, the multiple intermediate harmonics that can be determined are 200Hz and 300Hz. Based on this, the determined second target frequency bands are [50Hz, 150Hz], [150Hz, 250Hz], [250Hz, 350Hz], and [350Hz, 450Hz].

[0205] It is worth noting that each second target frequency band needs to cover the fundamental frequency, intermediate harmonics, or maximum harmonics.

[0206] Step 10224: Perform time-frequency domain conversion on the filtered previous vibration time series data and the current vibration time series data respectively, and obtain the vibration frequency domain features corresponding to each second target frequency band in the previous vibration time series data and the current vibration time series data respectively.

[0207] In this embodiment of the application, after determining multiple second target frequency bands containing effective signals of tool vibration based on the current tool's set rotation speed and number of cutting edges, the vibration frequency characteristics of the previous vibration time series data and the current vibration time series data in each second target frequency band can be obtained during time-frequency domain conversion, without needing to obtain vibration frequency domain characteristics of other frequency bands besides the second target frequency band, thereby improving the accuracy of the extracted vibration frequency domain characteristics.

[0208] Step 103: Based on the vibration feature matrix corresponding to the previous vibration time series data and the current vibration time series data, obtain the current feature difference degree between the previous machining process and the current machining process of the current tool.

[0209] In this embodiment, the difference in current characteristics between the current tool and the last two machining processes can be obtained by calculating the Euclidean distance between the vibration feature matrices corresponding to the previous vibration time series data and the current vibration time series data. Of course, the difference in current characteristics can also be obtained in other ways, and this application does not impose any particular limitations on this.

[0210] It is worth noting that before calculating the similarity of characteristics, the vibration time-domain features and vibration frequency-domain features in the vibration feature matrix can be standardized separately.

[0211] Step 104: Based on the comparison results of the current feature difference degree and the current difference degree threshold, determine the current state of the tool.

[0212] In this embodiment, the difference degree of the current feature between the current vibration time series data and the previous vibration time series data reflects the degree of change of the current vibration time series data compared to the previous vibration time series data. The smaller the difference degree of the current feature, the smaller the degree of change of the current vibration time series data compared to the previous vibration time series data; conversely, the larger the difference degree of the current feature, the greater the degree of change of the current vibration time series data compared to the previous vibration time series data. By comparing the above-mentioned difference degree of the current feature with the current difference degree threshold, the current tool state corresponding to the generation of the current vibration data is determined.

[0213] Please see Figure 4 This is a flowchart illustrating a method for determining the current tool state according to an embodiment of this application. Step 104 can be specifically implemented by executing sub-steps 1041 to 1043:

[0214] Step 1041: Determine whether the current feature difference is greater than the current difference threshold. Here, the current feature difference is the Euclidean distance between the vibration feature matrices corresponding to the previous vibration time series data and the current vibration time series data.

[0215] In this embodiment, the Euclidean distance is used to evaluate the difference in characteristics between the current vibration time series data and the vibration feature matrix corresponding to the previous vibration time series data. Therefore, the Euclidean distance between the current vibration time series data and the vibration feature matrix corresponding to the previous vibration time series data can be used to compare the set difference threshold.

[0216] Step 1042: If the difference is greater than the current difference threshold, then the current tool is determined to be in an abnormal state.

[0217] Step 1043: If the difference is not greater than the current difference threshold, then the tool is determined to be in normal condition.

[0218] In this embodiment, if the Euclidean distance between the current vibration timing data and the previous vibration timing data is large, it indicates that the current vibration timing data has changed significantly compared to the previous vibration timing data. If the tool state corresponding to the previous vibration timing data was normal, it can be inferred that the tool state was abnormal when the current vibration timing data was generated. If the tool state corresponding to the previous vibration timing data was abnormal, and if the Euclidean distance of the current tool is greater than the current difference threshold, it can be inferred that the tool state was still abnormal when the current vibration timing data was generated.

[0219] If the Euclidean distance between the current vibration timing data and the previous vibration timing data is small, it indicates that the change in the current vibration timing data compared to the previous vibration timing data is small. If the tool condition corresponding to the previous vibration timing data was normal, then it can be inferred that the tool condition when the current vibration timing data was generated was normal. If the tool condition corresponding to the previous vibration timing data was abnormal, and if the Euclidean distance of the current tool is not greater than the current difference threshold, then it can be inferred that the tool condition when the current vibration timing data was generated was still normal.

[0220] The following provides a detailed explanation of how to determine the current difference threshold and the initial difference threshold.

[0221] Please see Figure 5 This is a flowchart illustrating a method for determining a current difference threshold provided in an embodiment of this application. Before executing step 104, steps 201 to 204 may also be executed:

[0222] Step 201: Obtain the timing data of vibrations generated by the tool and the tool status data during the historical machining process of the CNC machine tool.

[0223] In this embodiment, vibration timing data generated by the tool within a preset time period and corresponding machining attribute information can be obtained. Specific machining attribute information can be found above and will not be repeated here. Then, the vibration timing data is filtered based on the machining attribute information. For example, vibration timing data not within a machining program segment is filtered out; vibration timing data with excessively long or short machining times is filtered out; vibration timing data where the machine tool is not in a machining state is filtered out; and vibration timing data where the machine tool's operating task is not material processing is filtered out, thus obtaining the vibration timing data generated by the tool during the machining process.

[0224] Then, the historical detection frequency, tool parameters, and alarm information of the tool setter can be obtained. Based on the number of vibration timing data points generated by the tool during the machining process and the detection frequency of the tool setter, the vibration timing data is grouped. For example, if the detection frequency is 5, a tool check is performed every 5 pieces of material processed. If the number of acquired vibration timing data points is 10, then the 10 vibration timing data points can be divided into 2 groups. For each group of vibration timing data, the tool setter performs a tool check after generating the last vibration data point in that group. If a broken tool is detected, an alarm message will be output.

[0225] For the vibration timing data of group n, if there is no alarm message after the last vibration timing data, the tool status corresponding to the vibration timing data of group n is determined to be normal. If there is an alarm message after the last vibration timing data, and the tool parameters indicate that the tool has not been replaced, the tool status corresponding to the vibration timing data of group n is determined to be suspected abnormal. If there is an alarm message after the last vibration timing data, and the tool parameters indicate that the tool has been replaced, the tool status corresponding to the last vibration timing data is determined to be abnormal, and the tool status corresponding to the other vibration timing data in group n, excluding the last vibration timing data, is determined to be suspected abnormal.

[0226] Step 202: Obtain the vibration feature matrix corresponding to the vibration time series data of each vibration.

[0227] In this embodiment of the application, the characteristics of each vibration time series data are extracted to obtain the corresponding vibration feature matrix. The vibration feature matrix includes vibration time domain features and vibration frequency domain features. For details on which vibration time domain features and vibration frequency domain features are included, please refer to the above description, which will not be repeated here.

[0228] It is worth noting that the tool status in each vibration time series data can be normal, abnormal, or suspected abnormal, and the appropriate status can be selected according to actual needs.

[0229] Step 203: Based on the vibration feature matrix corresponding to the vibration time series data, obtain the feature difference degree between two adjacent processing times in the historical processing of the CNC machine tool, and construct a set of feature difference degrees for each time.

[0230] In this embodiment, the feature difference degree can be obtained by calculating the Euclidean distance between the vibration feature matrices corresponding to adjacent vibration time series data, thereby constructing a set of feature difference degrees for each time series. Of course, other methods can also be used to obtain the feature difference degree, and this application does not impose any particular limitations on this.

[0231] It is worth noting that before calculating the feature difference, the vibration time-domain features and vibration frequency-domain features in the vibration feature matrix can be standardized separately. Since the main purpose of this application is to detect abnormal tool states, the vibration feature matrix corresponding to the tool state when it is normal is standardized only.

[0232] Step 204: With the aim of distinguishing between abnormal and normal tool states by the initial difference threshold, the initial difference threshold is iteratively adjusted based on the previous feature difference set and tool state until the iteration ends, and the initial difference threshold is obtained; the initial difference threshold is used as the initial value of the current difference threshold.

[0233] Two possible implementation schemes for threshold iterative adjustment are provided here.

[0234] For the first feature difference in each feature difference set, if this first feature difference is less than the initial value of the initial difference threshold, the predicted tool state is determined to be normal. Since both the actual and predicted tool states are normal at this time, it indicates that the initial value of the initial difference threshold is set appropriately. In this case, the initial difference threshold should be kept unchanged or updated in a decreasing direction. If the first feature difference is greater than the initial value of the initial difference threshold, the predicted tool state is determined to be abnormal. Since the actual tool state is normal while the predicted tool state is abnormal, it indicates that the initial value of the initial difference threshold is set too low. In this case, the initial difference threshold should be updated in a increasing direction. Subsequent iterations follow this pattern until the latest initial difference threshold is obtained.

[0235] The second approach involves using the current initial difference threshold to determine the tool status for n sets of vibration time series data, obtaining the judgment index (such as accuracy, false kill rate, etc.) corresponding to the n sets of vibration time series data, and then iteratively adjusting the initial difference threshold based on the judgment index.

[0236] In some embodiments, considering that the tool itself will be continuously worn during the machining process, the current difference threshold used to determine the tool condition should also be adaptively updated to improve the accuracy of tool condition determination.

[0237] Please see Figure 6 This is a flowchart illustrating a method for updating the current difference threshold provided in an embodiment of this application. Before executing step 104, steps 301 to 303 may also be executed:

[0238] Step 301: Obtain multiple vibration timing data of the processing process for the previously set number of times.

[0239] Step 302: Obtain the multiple vibration feature matrix corresponding to the multiple vibration time series data.

[0240] Step 303: Based on the multiple vibration time sequence data corresponding to the multiple vibration feature matrix, obtain the feature difference degree between two adjacent processing times in the previously set number of processing steps of the CNC machine tool.

[0241] Step 304: Update the current difference threshold based on the feature difference between two adjacent processing steps in the previously set number of processing steps.

[0242] In this embodiment of the application, due to the differences in the workpiece during the current tool processing, the current difference threshold may not be able to distinguish the current tool state well. Therefore, after each set number of processing operations, the current difference threshold will be updated to improve the accuracy of tool state judgment.

[0243] The following section provides a detailed explanation of how to update the current difference threshold.

[0244] Please see Figure 7 This is a flowchart illustrating the method for updating the current difference threshold provided in this application embodiment. Step 304 can be specifically implemented by executing sub-steps 3041 to 3044:

[0245] Step 3041: Obtain the feature difference degree corresponding to two adjacent times in the previous processing, and construct the difference degree set.

[0246] Step 3042: Obtain the threshold to be updated based on the mean and variance of the feature differences corresponding to the difference set, and the current tool size parameters.

[0247] Step 3043: Determine whether the threshold to be updated is greater than the initial difference threshold.

[0248] Step 3044: If it is greater than the threshold to be updated, then use the threshold to be updated as the current difference threshold; if it is less than the threshold, then use the initial difference threshold as the current difference threshold.

[0249] In this embodiment, for several previous processing steps, the mean and variance of overall similarity corresponding to multiple feature differences can be determined based on the feature differences between the vibration time sequence data generated by two adjacent processing steps. Then, the threshold to be updated is determined based on the mean, variance, and current tool size parameters. The calculation formula for the threshold to be updated is shown in (28):

[0250] y=μ+cl*σ(28)

[0251] Where y represents the threshold to be updated, μ represents the mean of the feature differences corresponding to the difference set, σ represents the variance of the feature differences corresponding to the difference set, and cl represents the tool size parameter. Of course, in the above embodiments, the standard deviation can also be used instead of the variance, and this application does not impose any particular restriction on this.

[0252] If the threshold to be updated is greater than the initial difference threshold obtained from previous training, then the threshold to be updated is used as the new current difference threshold. It should be understood that the tool anomaly judgment mechanism in this application is based on the current feature difference being greater than the current difference threshold. Therefore, if the threshold to be updated is greater than the initial difference threshold obtained from previous training, the threshold to be updated is used as the new current difference threshold. This means that the current feature difference needs to be greater than the threshold to be updated to be judged as a tool anomaly. Here, a larger current difference threshold is used as the standard for judging tool anomalies, modifying the detection adaptability to better detect tool status.

[0253] If the threshold to be updated is not greater than the initial difference threshold obtained from previous training, then the initial difference threshold is used as the new current difference threshold, that is, the lower limit of the threshold is used to detect the tool status.

[0254] Please see Figure 8 This application provides an electronic device, which includes at least one processor 401. The processor 401 is used to execute a computer program stored in a memory to implement the functionality provided in this application embodiment. Figure 1-7 The flowchart illustrates the steps of the method for detecting the tool status of a CNC machine tool.

[0255] Optionally, the processor 401 may be a central processing unit, a specific ASIC, or one or more integrated circuits used to control program execution.

[0256] Optionally, the electronic device may further include a memory 402 connected to at least one processor 401. The memory 402 may include ROM, RAM, and disk storage. The memory 402 stores data required for the processor 401 to run, i.e., it stores instructions that can be executed by at least one processor 401. The at least one processor 401 executes instructions stored in the memory 402 to perform tasks such as... Figure 1-7 The method is shown. The number of memories 402 is one or more.

[0257] This application embodiment also provides a computer storage medium, wherein the computer storage medium stores a computer program, and when the computer program runs on a processor, it causes the processor to perform actions such as... Figure 1-7 The method.

[0258] The above description is merely a preferred embodiment of this specification and is not intended to limit this specification. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this specification should be included within the scope of protection of this specification.

Claims

1. A method for detecting the cutting tool condition of a CNC machine tool, characterized in that, The method includes: Acquire the previous vibration timing data and the current vibration timing data generated by the current tool during the machining process of the CNC machine tool; Based on the previous vibration time series data and the current vibration time series data, vibration feature matrices corresponding to the previous vibration time series data and the current vibration time series data are obtained respectively; wherein, the vibration feature matrix includes vibration time domain features and vibration frequency domain features; Based on the vibration feature matrix corresponding to the previous vibration time sequence data and the current vibration time sequence data, the current feature difference degree of the current tool between the previous processing and the current processing is obtained. Based on the comparison results between the current feature difference degree and the current difference degree threshold, the current state of the current tool is determined.

2. The detection method according to claim 1, characterized in that, The step of obtaining vibration feature matrices corresponding to the previous vibration time series data and the current vibration time series data, respectively, based on the previous vibration time series data and the current vibration time series data, includes: The previous vibration time series data and the current vibration time series data are filtered based on the first target frequency band; Based on the filtered previous vibration time series data and the current vibration time series data, the vibration time domain features and vibration frequency domain features corresponding to the previous vibration time series data and the current vibration time series data are extracted respectively.

3. The detection method according to claim 2, characterized in that, Based on the filtered previous vibration time-series data and the current vibration time-series data, the vibration time-domain features and vibration frequency-domain features corresponding to the previous vibration time-series data and the current vibration time-series data are extracted respectively, including: The fundamental frequency for time-frequency domain conversion is determined based on the current set rotational speed of the tool. Based on the current number of cutting edges of the tool and the fundamental frequency, determine the maximum harmonic of the time-frequency domain conversion; Based on the fundamental frequency and the maximum harmonic, at least two second target frequency bands for time-frequency domain conversion are determined; The filtered previous vibration time series data and the current vibration time series data are respectively converted into time and frequency domains to obtain the vibration frequency domain features corresponding to each second target frequency band in the previous vibration time series data and the current vibration time series data.

4. The detection method according to claim 1, characterized in that, The determination of the current state of the tool based on the comparison result between the current feature difference degree and the current difference degree threshold includes: Determine whether the current feature difference is greater than the current difference threshold; wherein, the current feature difference is the Euclidean distance between the vibration feature matrices corresponding to the previous vibration time series data and the current vibration time series data; If the difference is greater than the current difference threshold, then the current tool is determined to be in an abnormal state.

5. The detection method according to claim 4, characterized in that, After determining whether the current feature difference is greater than the current difference threshold, the method further includes: If the difference is not greater than the current difference threshold, then the current tool is determined to be in a normal state.

6. The detection method according to claim 1, characterized in that, Before determining the current state of the tool based on the comparison result between the current feature difference degree and the current difference degree threshold, the method further includes: Acquire multiple vibration timing data of the processing procedure with a previously set number of cycles; Obtain the multiple vibration feature matrix corresponding to the multiple vibration time series data; Based on the vibration feature matrix corresponding to the multiple vibration time series data, the feature difference degree between two adjacent processing operations in the previous set number of processing operations of the CNC machine tool is obtained; The current difference threshold is updated based on the feature difference between two adjacent processing operations during the previously set number of processing operations.

7. The detection method according to claim 6, characterized in that, The step of updating the current difference threshold based on the comparison results of the feature difference between two adjacent processing steps and the initial threshold includes: Obtain the feature difference between two adjacent processing steps in the previous processing steps, and construct a difference set; Based on the mean and variance of the feature differences corresponding to the difference set, and the size parameters of the current tool, obtain the threshold to be updated; Determine whether the threshold to be updated is greater than the initial difference threshold; If it is greater than the threshold to be updated, then the threshold to be updated will be used as the current difference threshold. If it is not greater than, then the initial difference threshold is used as the current difference threshold.

8. The detection method according to claim 7, characterized in that, Before determining the current state of the tool based on the comparison result between the current feature difference degree and the current difference degree threshold, the method further includes: Acquire the historical vibration timing data and tool status data generated by the tool during the machining process of the CNC machine tool. Obtain the vibration feature matrix corresponding to the vibration time series data of each vibration; Based on the vibration feature matrix corresponding to the vibration time series data, the feature difference degree corresponding to two adjacent processing times in the historical processing of the CNC machine tool is obtained, and a set of feature difference degrees for each time is constructed. With the aim of distinguishing between abnormal and normal tool states by the initial difference threshold, the initial difference threshold is iteratively adjusted based on the previous feature difference set and the tool state until the iteration ends, and the initial difference threshold is obtained. The initial difference threshold is used as the initial value of the current difference threshold.

9. An electronic device, characterized in that, Includes memory and processor, wherein: The memory is used to store computer programs; The processor is configured to execute the computer program to implement the detection method as described in any one of claims 1 to 8.

10. A computer-readable storage medium, characterized in that, Used to store a computer program, wherein the computer program, when executed by a processor, implements the detection method as described in any one of claims 1 to 8.

Images