Chip removal control method and system for machining center

Abstract

The invention belongs to the technical field of metal cutting machining control, and particularly relates to a chip removal control method and system for a machining center, and the method comprises the steps: collecting and preprocessing a real-time vibration signal of the machining center, and obtaining a pure vibration signal; performing periodic stability analysis on the pure vibration signal, obtaining the periodic stability purity of each frequency band signal, and determining an optimal characteristic frequency band; extracting features capable of representing the chip accumulation evolution trend based on the optimal feature frequency band, and grading the chip accumulation state in real time; and executing a matched active avoidance or grading response control strategy according to the chip accumulation state grade. According to the method, random chip accumulation impact and normal machining load fluctuation can be accurately distinguished, the monitoring robustness and sensitivity are remarkably improved, and intelligent control upgrading from passive response to active avoidance is achieved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of metal cutting process control. More particularly, the present application relates to a method and system for chip control of machining center. BACKGROUND

[0002] In the process of metal cutting, especially deep hole or complex cavity machining, a large amount of chips will be generated. If these chips cannot be effectively removed in time, they will accumulate and entangle around the tool, and even be subjected to secondary cutting by the tool. This chip accumulation phenomenon not only seriously scratches the machined surface of the workpiece, leading to surface roughness exceeding the standard and reducing the machining precision, but also can cause excessive wear or sudden collapse of the tool. In the process of high-precision machining, even a small amount of chip accumulation can cause the workpiece to be scrapped, resulting in huge economic losses. In the automatic production line, severe chip accumulation can force the production to be interrupted, affecting the overall production efficiency, and when the chip accumulation reaches a certain degree, it can cause safety accidents, posing a threat to the safety of operators and equipment. Therefore, early warning and effective control of chip accumulation are of great significance to ensure machining quality, improve production efficiency and ensure production safety.

[0003] In order to realize early warning of chip accumulation, the prior art proposes to analyze the vibration signal, but most of them rely on static and apparent feature analysis of signal form, such as kurtosis, sparsity, waveform factor and other time domain statistical characteristics. This method is essentially a blind source processing, that is, without understanding the signal generation mechanism, only the external characteristics of the signal are extracted and recognized, and only the amplitude or energy level is analyzed, which cannot effectively distinguish between normal cutting vibration and random impact caused by chip accumulation. In complex working conditions, such as tool cutting into / out of the workpiece, uneven material or process parameter change, the vibration signal will produce similar changes as chip accumulation, resulting in high false alarm rate of the prior art. SUMMARY

[0004] In order to solve the technical problems that the prior art cannot effectively distinguish between normal cutting vibration signal and random impact signal caused by chip accumulation, resulting in insufficient monitoring sensitivity and robustness, the present application provides solutions in the following aspects.

[0005] In a first aspect, the present application provides a chip control method for machining center, comprising: The real-time vibration signal of the machining center is collected, the real-time vibration signal is preprocessed to obtain a pure vibration signal, the pure vibration signal is converted to a frequency domain, the frequency domain is divided into a plurality of equal-width preselected frequency bands, a cyclic spectrum density function of each preselected frequency band is calculated, a periodic stationary purity index of each preselected frequency band is determined according to the cyclic spectrum density function, an optimal feature frequency band is determined according to the periodic stationary purity index, a feature signal in the time domain is extracted from the optimal feature frequency band, a chip accumulation evolution trend index at a current moment is determined according to the feature signal, a chip accumulation state is classified according to a change of the chip accumulation evolution trend index, and a control strategy matched with the classified chip accumulation state is executed according to the classified chip accumulation state.

[0006] The present application can accurately capture the damage to the normal cutting periodicity caused by chip accumulation by analyzing the periodic stationary property of each preselected frequency band, starting from the internal rhythm and dynamic characteristics of the signal, thereby identifying early abnormalities with high sensitivity, having stronger anti-interference ability compared with the traditional analysis method relying on signal amplitude or energy form, being able to effectively distinguish normal machining load fluctuations from real random abnormalities, and significantly reducing the false alarm rate.

[0007] Preferably, the preprocessing is noise reduction processing.

[0008] Preferably, the periodic stationary purity index satisfies the expression: ; in the expression, is a periodic stationary purity index of the i-th preselected frequency band; is a periodic stationary purity index of the i-th preselected frequency band; is a cyclic spectrum density function of the i-th preselected frequency band signal; is a cyclic spectrum density function of the i-th preselected frequency band signal; is a normal frequency; is a cyclic frequency; is the i-th feature cyclic frequency related to normal cutting, and a set of the i-th feature cyclic frequency is composed of a tooth pass frequency and a harmonic thereof; is the i-th feature cyclic frequency related to normal cutting, and a set of the i-th feature cyclic frequency is composed of a tooth pass frequency and a harmonic thereof; is the total number of the selected feature cyclic frequencies.

[0009] The periodic stationary purity index in the present application reflects the energy proportion of the periodic stationary component, when random impacts are introduced due to chip accumulation, the non-periodic energy will significantly increase, thereby causing the periodic stationary purity index to decrease, and the periodic stationary purity index can reflect the purity degree of the periodicity of the signal, thereby providing a reliable data basis for identifying chip accumulation.

[0010] Preferably, the determination of the optimal feature frequency band according to the periodic stationary purity index comprises: taking the preselected frequency band with the minimum periodic stationary purity index as the optimal feature frequency band.

[0011] The lower the cyclostationary purity index of the frequency band, the more sensitive the frequency band is to the chip accumulation phenomenon, and the optimal feature frequency band with the minimum cyclostationary purity index is selected in the application, so that the signal frequency band that can best reflect the early chip accumulation characteristics is adaptively locked, and the optimal signal input is provided for subsequent fine feature extraction and state judgment.

[0012] Preferably, the feature signal in the time domain is extracted from the optimal feature frequency band, comprising: performing wavelet packet decomposition on the pure vibration signal according to the center frequency and bandwidth of the optimal feature frequency band, and reconstructing the time domain signal corresponding to the optimal feature frequency band as the feature signal.

[0013] Preferably, the chip accumulation evolution trend index satisfies the expression: ; in the expression, is the serial number of the current time, is the chip accumulation evolution trend index of the current time; is the adaptive smoothing factor of the current time; is the chip accumulation evolution trend index of the previous time.

[0014] The application converts a series of instantaneous vibration information into a continuous chip accumulation evolution trend index that can stably reflect the accumulation process and severity, effectively smooths the short-term fluctuations of the signal, and makes it better reflect the long-term development trend of chip accumulation, thereby providing a reliable decision basis for subsequent state classification.

[0015] Preferably, the adaptive smoothing factor satisfies the expression, comprising: ; in the expression, is the serial number of the current time, is the adaptive smoothing factor of the current time; is the amplitude of the feature signal of the current time; is the average amplitude baseline of the feature signal learned in the normal cutting stage without chip accumulation; is the standard deviation of the feature signal in the normal cutting stage; is a natural exponential function; and is the minimum and maximum value boundary of the smoothing factor; is the sensitivity coefficient.

[0016] The application realizes adaptive adjustment of the smoothing factor by calculating the normalized degree of deviation of the current signal point from the normal baseline, and when the signal fluctuates within the normal range, the adaptive smoothing factor tends to the minimum value, and the system exhibits strong memory; when the signal deviates greatly from the baseline, the adaptive smoothing factor tends to the maximum value, and the system exhibits strong sensitivity, so that the system can switch between stability and response speed, and ensure accurate tracking of the evolution process of the chip accumulation.

[0017] Preferably, the grading of the chip accumulation state comprises: determining a normal cutting state in response to the chip accumulation evolution trend indicator at the current moment being less than an alarm threshold ; determining a trend growth state in response to the chip accumulation evolution trend indicator at the current moment being greater than or equal to the alarm threshold , and the growth rate of the chip accumulation evolution trend indicator at the current moment being less than a growth rate threshold; determining a sharp deterioration state in response to the chip accumulation evolution trend indicator at the current moment being greater than or equal to the alarm threshold , and the growth rate of the chip accumulation evolution trend indicator at the current moment being greater than or equal to the growth rate threshold; and determining a serious accumulation state in response to the chip accumulation evolution trend indicator at the current moment being greater than or equal to , and lasting for more than milliseconds, wherein is a serious coefficient greater than .

[0018] Preferably, the control strategy matched with the grade of the chip accumulation state determined in real time is executed, comprising: executing a silent monitoring strategy in response to the normal cutting state; executing a low-intensity early warning type chip removal strategy in response to the trend growth state; executing a real-time machining parameter intervention strategy in response to the sharp deterioration state; and executing a protective shutdown and strong chip removal strategy in response to the serious accumulation state.

[0019] In a second aspect, the application provides a chip removal control system for a machining center, comprising a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the above-mentioned chip removal control method for the machining center is realized.

[0020] By using the above technical solution, the above-mentioned chip removal control method for the machining center is generated into a computer program and stored in the memory to be loaded and executed by the processor, so that a terminal device is manufactured according to the memory and the processor, and use is facilitated.

[0021] The beneficial effects of the present application are that: the present application can accurately capture the damage to the normal cutting periodicity caused by chip accumulation by constructing the periodicity stability purity index, starting from the internal rhythm and dynamic characteristics of the signal, thereby identifying early abnormalities with high sensitivity, and has stronger anti-interference ability compared with traditional analysis methods relying on signal amplitude or energy form; the present application introduces a real-time machining parameter intervention strategy matched with the accumulation state level, so that the system upgrades from a passive response mode of cleaning up after the problem occurs to an active avoidance mode of preventing the problem from worsening from the root of the process, effectively ensuring the continuity and stability of the machining process, avoiding downtime caused by serious accumulation, avoiding unnecessary powerful chip removal, reducing the consumption of cooling liquid and compressed air, and achieving comprehensive optimization of production quality, efficiency and cost. BRIEF DESCRIPTION OF DRAWINGS

[0022] Figure 1 is a flow chart schematically showing a chip removal control method for a machining center in the present application; Figure 2 is a schematic diagram of a pure vibration signal; Figure 3 is a schematic diagram of optimal feature band acquisition; Figure 4 is a schematic diagram of chip accumulation evolution trend index change. DETAILED DESCRIPTION

[0023] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0024] The specific embodiments of the present application will be described in detail below with reference to the drawings.

[0025] The embodiments of the present application disclose a chip removal control method for a machining center, referring to Figure 1 , comprising steps S1-S4: S1, collecting real-time vibration signals of the machining center, pre-processing the real-time vibration signals to obtain pure vibration signals.

[0026] It should be noted that when precise machining is performed on the machining center, any slight abnormality, such as normal scraping of a single chip with a tool, will leave a trace in the overall vibration of the machine tool. However, these key weak signals are often masked by strong background noise generated by the main shaft rotation, cooling liquid pump, etc. Therefore, the present application collects real-time vibration signals of the machining center and performs noise reduction processing on the vibration signals.

[0027] Specifically, the acceleration sensor is installed near the spindle bearing seat or tool holder of the machining center to collect real-time vibration signals of the machining center.

[0028] In the embodiment, the sampling frequency is set to , wherein is the maximum spindle speed that the machining center can reach, in rpm, is the maximum number of teeth of the tool that can be used, for example, when rpm, , the sampling frequency is Hz. In other embodiments, the implementer can adjust the sampling frequency according to the parameters of the actual machining center, but should ensure that the sampling frequency is not less than 10 times the maximum tooth pass frequency of the tool to fully capture the vibration characteristics in the cutting process.

[0029] Further, the collected original vibration signals are denoised to obtain pure vibration signals. In the embodiment, the original vibration signals are denoised by median filtering to eliminate the interference of environmental and electrical noise. It should be noted that in the deep hole drilling process, median filtering is particularly suitable for processing pulse-type noise, which can effectively retain the transient impact characteristics generated by chip accumulation, while suppressing the periodic interference generated by the cooling liquid pump and the like. In other embodiments, the implementer can also select the denoising algorithm according to the actual implementation, for example, wavelet threshold denoising method. Exemplarily, Figure 2 is a schematic diagram of the pure vibration signal.

[0030] S2, convert the pure vibration signal to the frequency domain, divide the frequency domain into several equal-width preselected frequency bands, calculate the cyclic spectral density function of each preselected frequency band, determine the periodic stationary purity index of each preselected frequency band according to the cyclic spectral density function, and determine the optimal feature frequency band according to the periodic stationary purity index.

[0031] It should be noted that the rotational motion of the tool of the numerical control machining center causes the vibration signals generated by normal cutting to have the characteristic of periodic change over time in statistical characteristics, and the periodic rhythm is strictly synchronized with the spindle speed, tool tooth number and other process parameters. The random collision and scraping caused by early chip accumulation is a non-stationary and non-periodic impact interference. Such interference can destroy the periodic stationarity of the original signal. Therefore, the present application analyzes the purity of the periodic stationarity of each frequency band signal to inversely identify the strength of the non-periodic interference, thereby identifying the most sensitive feature frequency band.

[0032] Specifically, the spindle speed and the number of teeth of the currently used tool are obtained in real time from the numerical control system of the machining center, and the fundamental frequency component related to the normal cutting process, i.e., the tooth pass frequency and its harmonic frequencies are calculated wherein is a positive integer.

[0033] Further, the pre-processed pure vibration signal is subjected to time-frequency domain transformation, and the entire frequency domain is divided into equal-width pre-selected frequency bands, wherein = f s / N , is a sampling frequency, is a gear pass frequency, is a down-rounding symbol. The present application sets the frequency band width to of the gear pass frequency, which can ensure sufficient frequency band resolution around each harmonic frequency, thereby accurately capturing the change of periodic components.

[0034] The cyclic spectral density function of the signal in each pre-selected frequency band is calculated.

[0035] It should be noted that the cyclic spectral density function is a prior art in the field of signal processing, which is used to analyze the frequency domain characteristics of a periodic stationary signal. In the present embodiment, the time-domain smooth periodogram method is used to calculate the cyclic spectral density function, and the specific process is as follows: first, the cyclic autocorrelation function of the signal is calculated, and then the Fourier transform of the cyclic autocorrelation function is performed to obtain the cyclic spectral density function. In the calculation process, the signal is subjected to windowing processing by using a Hanning window to reduce spectral leakage. The cyclic spectral density function can exhibit the distribution of signal energy on the conventional frequency axis and the cyclic frequency axis .

[0036] Further, according to the cyclic spectral density function, the periodic stationary purity index of each pre-selected frequency band is calculated:

[0037] wherein is the periodic stationary purity index of the th pre-selected frequency band; is the cyclic spectral density function of the th pre-selected frequency band signal; is a conventional frequency; is a cyclic frequency; is the th characteristic cyclic frequency related to normal cutting, and the set thereof is composed of the gear pass frequency and its harmonics, i.e. ; is the total number of the selected characteristic cyclic frequencies, in order to balance the reliability of detection and the efficiency of real-time calculation, the value range of is usually between 2 and 5, and the experienced value is 3. The implementer can set the value of according to the actual implementation situation.

[0038] It should be noted that, in order to quantify the relative intensity of regular and random components in a signal, this invention constructs a periodic stationary purity index. The numerator of the periodic stationary purity index is calculated by integration at the characteristic cycle frequency of the signal energy related to regular normal cutting. The sum of the numerator and denominator represents the energy of the periodic stationary components in the signal; the denominator is the sum of the energies of the periodic stationary components and the non-periodic stationary components, representing the overall energy of the signal. Therefore, the periodic stationary purity index reflects the proportion of the energy of the periodic stationary components to the total energy. During normal cutting, the signal is highly periodic, with most of the energy concentrated at the characteristic cycle frequency, and the numerator is approximately equal to the denominator. The value approaches When chips begin to accumulate, the non-periodic collisions between the tool and the chips generate random impacts. These impacts disrupt the periodic vibrations produced by normal cutting, affecting various positions in the cycle spectrum, especially... The axis introduces a large amount of aperiodic energy, causing the energy of the aperiodic stationary component in the denominator to increase, while the numerator remains essentially unchanged or even decreases, thus leading to... The value decreases significantly. Therefore, the smaller the periodic stability purity index, the more severe the influence of random shocks on the preselected frequency band, and the more sensitive it is to chip accumulation.

[0039] Furthermore, the pre-selected frequency band with the smallest periodic stability purity index is taken as the optimal characteristic frequency band, and the center frequency of this optimal characteristic frequency band is taken as the optimal center frequency, and its bandwidth is taken as the optimal bandwidth. For example, Figure 3 This is a schematic diagram for obtaining the optimal feature frequency band.

[0040] S3. Extract the time-domain feature signal from the optimal feature frequency band, and determine the chip accumulation evolution trend index at the current moment based on the feature signal.

[0041] Specifically, based on the optimal center frequency and optimal bandwidth, a customized wavelet packet decomposition is performed on the pure vibration signal, and the time-domain signal corresponding to the optimal characteristic frequency band is reconstructed as the characteristic signal.

[0042] The number of wavelet packet decomposition layers is determined based on the optimal center frequency and the sampling frequency.

[0043] In the formula, The number of wavelet packet decomposition levels. For the optimal center frequency, Sampling frequency, This is a floor function. This invention selects the db4 wavelet from the Daubechies wavelet family as the basis function, and decomposes the signal into its constituent parts through wavelet packet decomposition. Each frequency band, and then based on the optimal center frequency and optimal bandwidth The corresponding frequency band coefficient is selected for reconstruction to obtain a time domain signal corresponding to the optimal characteristic frequency band. It should be noted that the db4 wavelet in the Daubechies wavelet system is selected as the base function in the embodiment, because the db4 wavelet has good time-frequency localization characteristics and is suitable for analyzing transient impact signals in the cutting process. In other embodiments, the implementer can select the base function for wavelet packet decomposition according to the actual implementation situation.

[0044] Further, according to the statistical characteristics of the characteristic signal, a smoothing factor for calculating the evolution trend index is determined:

[0045] In the formula, is the serial number of the current time, is the adaptive smoothing factor of the current time; is the amplitude of the characteristic signal at the current time; is the average amplitude baseline of the characteristic signal learned in the normal cutting stage without chip accumulation; is the standard deviation of the characteristic signal in the normal cutting stage; is a natural exponential function.

[0046] In the formula, and are the minimum and maximum value boundaries of the smoothing factor, and the empirical values are 0.05 and 0.95. In other embodiments, the implementer can adjust and according to the actual machining conditions. For example, in high-precision machining, to improve system stability and reduce false positives, the value of may be increased to 0.1, may be reduced to 0.9; in rough machining or working conditions prone to chip accumulation, to improve detection sensitivity and avoid false negatives, the value of may be reduced to 0.01, may be increased to 0.99. It should be noted that should not be set too small, generally not less than 0.01, otherwise the system will be too sensitive in a stable state and easily disturbed by random noise; should not be set too large, generally not higher than 0.99, otherwise the system will have weak memory in a sudden state and will not be able to effectively track the development trend of the anomaly.

[0047] In the formula, is a sensitivity coefficient for adjusting the threshold for distinguishing between normal fluctuations and abnormal signals. According to statistical experience rules, about 99.7% of normal data will fall within the range of mean ± 3 standard deviations, so the value of is set to 3, the implementer can also set it according to the actual implementation, when The larger the setting is, the higher the threshold for the system to determine as an anomaly is, and the lower the false positive rate is but the early chip accumulation phenomenon may be missed, when The smaller the setting is, the higher the sensitivity of the system to the anomaly is, and the initial sign of chip accumulation can be detected earlier, but the false positive rate is increased accordingly. In actual application, the implementer should weigh according to the processing process requirement, false positive tolerance and the risk of missing, for example, in high-precision processing, the value of may be increased appropriately to reduce the false positive rate, while in rough machining or under the working condition prone to chip accumulation, the value of may be reduced appropriately to improve the detection sensitivity.

[0048] In the formula, The degree of deviation of the characteristic signal at the current time from the normal baseline is calculated, and the deviation degree of the signal is mapped into the value range of the smoothing factor by an exponential decay function in the present application, so as to realize the adaptive adjustment of the smoothing factor. When the system is in the normal cutting stage without chip accumulation, the signal fluctuation is small, the deviation degree is small, when the deviation degree is less than the sensitivity coefficient , the exponential term tends to 1, so that tends to , at this time tends to the minimum value , the system shows strong memory and can effectively filter random noise; on the contrary, in the early stage of chip accumulation, the signal produces mutation, the signal point deviates from the baseline greatly, when the deviation degree exceeds , the exponential term tends to , so that tends to , at this time tends to the maximum value , the system can quickly improve the sensitivity and timely capture the development trend of the anomaly, so as to provide sufficient time window for subsequent active intervention.

[0049] Further, according to the characteristic signal and the adaptive smoothing factor, the chip accumulation evolution trend index is calculated:

[0050] In the formula, is the serial number at the current time, is the chip accumulation evolution trend index at the current time; is the adaptive smoothing factor at the current time; is the chip accumulation evolution trend index of the last time. Figure 4 is the chip accumulation evolution trend index change schematic diagram.

[0051] S4, according to the change of the chip accumulation evolution trend index, the chip accumulation state is graded, according to the real-time determined chip accumulation state grade, the control strategy matched with the grade is executed.

[0052] Specifically, the dynamic adaptive alarm threshold matched with the working condition is set The acquisition method is as follows: at the initial stage of each new working condition, the vibration signal under the non-accumulation state is collected, the mean value of the chip accumulation evolution trend index in this stage is calculated And the standard deviation And the alarm threshold Is set as: .

[0053] Further, the growth rate of the chip accumulation evolution trend index at the current time is calculated : , wherein is the chip accumulation evolution trend index at the current time, is the chip accumulation evolution trend index of the last time, is the sampling time interval.

[0054] The growth rate threshold : through the data analysis of historical chip accumulation events, the mean value And the standard deviation Of the growth rate In the normal cutting state are calculated, and then , wherein is the safety factor, which can be adjusted by the implementer according to the actual machining condition , for example, in high-precision machining, in order to avoid misjudgment, the Can be increased to 0.8, so that the growth rate threshold Increases; in rough machining, in order to improve the detection sensitivity, the Can be reduced to 0.3, so that the growth rate threshold Reduces.

[0055] In response to the chip accumulation evolution trend index at the current time Is less than the alarm threshold , it is determined that the normal cutting state; in response to the chip accumulation evolution trend index at the current time greater than or equal to the alarm threshold , and the growth rate of the chip accumulation evolution trend index at the current moment less than the growth rate threshold , the trend growth state is determined; in response to the chip accumulation evolution trend index at the current moment greater than or equal to the alarm threshold , and the growth rate of the chip accumulation evolution trend index at the current moment greater than or equal to the growth rate threshold , the sharp deterioration state is determined; in response to the chip accumulation evolution trend index at the current moment greater than or equal to , and lasting for milliseconds or more, the serious accumulation state is determined. Wherein, is a serious coefficient greater than , which is set by the implementer according to the actual implementation, for example .

[0056] It should be noted that the duration threshold of milliseconds is set to avoid misjudgment caused by temporary fluctuations and to ensure that only serious abnormalities that persist will trigger the highest level of response. In actual machining process, when the system detects the chip accumulation trend, if strong chip removal measures are taken immediately, it may interfere with the normal machining process and affect the surface quality. Through the hierarchical response strategy, the system can take low-intensity intervention in the early stage of chip accumulation and eliminate hidden dangers without interrupting the machining process. Only when the accumulation is serious and threatens the machining quality, will strong measures such as stopping be taken, thereby maximizing the continuity and stability of the machining process.

[0057] Specifically, in response to the normal cutting state, a silent monitoring strategy is executed, and the system does not intervene; in response to the trend growth state, a low-intensity early warning chip removal strategy is executed, instructions are sent to the PLC to trigger a short and targeted pulse jet, or the coolant flow is temporarily and slightly increased and then restored, to eliminate early hidden dangers at the lowest cost. In actual deep hole drilling process, short and targeted pulse jet can effectively remove the just beginning to accumulate chips without interfering with the normal drilling process; in response to the sharp deterioration state, a real-time machining parameter intervention strategy is executed, instructions are preferentially sent to the CNC controller to actively avoid problem deterioration; in response to the serious accumulation state, a protective shutdown and strong chip removal strategy is executed, a "feed hold" instruction is sent to the CNC to pause the machining, and instructions are sent to the PLC to trigger the highest power strong chip removal program and issue an alarm signal.

[0058] Optionally, in the case of sharp deterioration, the processing parameter intervention strategy includes at least one of the following: 1. Dynamic adjustment of feed rate: instantaneously reduce the current feed rate to reduce the amount of chips generated per unit time.

[0059] 2. Spindle speed perturbation: within the current speed range, periodic fluctuations with a frequency of Hz, using harmonic changes to assist in breaking long chips.

[0060] 3. Instantaneous adjustment of tool path: in cavity machining, control the tool to temporarily retreat a small amount in the axial direction mm and then recover, using the gap to allow coolant to flow in and flush the chips.

[0061] It should be noted that in the process of complex cavity milling, when the chip accumulation deteriorates sharply, the spindle speed perturbation strategy can effectively utilize the resonance effect to break long chips and avoid their winding around the tool; and the instantaneous adjustment of the tool path can create opportunities for the coolant to enter the tool-workpiece interface without interrupting the machining process, effectively flushing the accumulated chips.

[0062] The embodiment of the application also discloses a chip removal control system for a machining center, comprising a processor and a memory, the memory storing computer program instructions, which, when executed by the processor, realize a chip removal control method for a machining center according to the application.

[0063] The above system also includes a communication bus and a communication interface and other components familiar to those skilled in the art, the settings and functions of which are known in the art, and therefore will not be described here.

Claims

1. A chip removal control method for a machining center, characterized in that, include: The real-time vibration signal of the machining center is collected and preprocessed to obtain a clean vibration signal. The pure vibration signal is converted to the frequency domain, the frequency domain is divided into several pre-selected frequency bands of equal width, the cyclic spectral density function of each pre-selected frequency band is calculated, the periodic stationary purity index of each pre-selected frequency band is determined based on the cyclic spectral density function, and the optimal characteristic frequency band is determined based on the periodic stationary purity index. Extract time-domain feature signals from the optimal feature frequency band, and determine the chip accumulation evolution trend index at the current moment based on the feature signals; The chip accumulation state is classified according to the changes in the chip accumulation evolution trend index; and a control strategy matching the real-time determined chip accumulation state level is executed.

2. The chip removal control method for a machining center according to claim 1, characterized in that, The preprocessing is noise reduction.

3. The chip removal control method for a machining center according to claim 1, characterized in that, The periodic stable purity index satisfies the expression: ； In the formula, For the first The periodic stability purity index of each pre-selected frequency band; For the first Cyclic spectral density function of a preselected frequency band signal; This is a standard frequency; The cycle frequency; For the first related to normal cutting A set of characteristic cyclic frequencies, which consists of the tooth frequency and its harmonics; This represents the total number of selected characteristic cycle frequencies.

4. The chip removal control method for a machining center according to claim 1, characterized in that, The determination of the optimal characteristic frequency band based on the periodic stable purity index includes: The pre-selected frequency band with the smallest periodic stability purity index is taken as the optimal feature frequency band.

5. A chip removal control method for a machining center according to claim 1, characterized in that, Extracting the time-domain feature signal from the optimal feature frequency band includes: Based on the center frequency and bandwidth of the optimal characteristic frequency band, wavelet packet decomposition is performed on the pure vibration signal to reconstruct the time-domain signal corresponding to the optimal characteristic frequency band, which is then used as the characteristic signal.

6. The chip removal control method for a machining center according to claim 1, characterized in that, The chip accumulation evolution trend index satisfies the following expression: ； In the formula, This is the sequence number of the current time. This serves as an indicator of the current chip accumulation evolution trend. The adaptive smoothing factor at the current moment; This is an indicator of the chip accumulation evolution trend at the previous moment.

7. A chip removal control method for a machining center according to claim 6, characterized in that, The adaptive smoothing factor satisfies the following expression: ； In the formula, This is the sequence number of the current time. The adaptive smoothing factor at the current moment; The characteristic signal amplitude at the current moment; The baseline of the average amplitude of the feature signal learned during the normal cutting phase without chip buildup; The standard deviation of the characteristic signal during the normal cutting stage; It is a natural exponential function; and These are the minimum and maximum value boundaries of the smoothing factor; This is the sensitivity coefficient.

8. A chip removal control method for a machining center according to claim 1, characterized in that, The classification of chip accumulation state includes: The response is that the chip accumulation evolution trend index at the current moment is less than the alarm threshold. When the chip accumulation evolution trend index at the current moment is greater than or equal to the alarm threshold, it is determined to be a normal cutting state. When the growth rate of the chip accumulation evolution trend index at the current moment is less than the growth rate threshold, it is determined to be in a trend growth state; in response, when the chip accumulation evolution trend index at the current moment is greater than or equal to the alarm threshold... When the growth rate of the chip accumulation evolution trend index at the current moment is greater than or equal to the growth rate threshold, it is judged to be in a state of rapid deterioration; in response to the chip accumulation evolution trend index at the current moment being greater than or equal to... and continue When the time exceeds milliseconds, it is considered a severe accumulation state, where, greater than The severity coefficient.

9. A chip removal control method for a machining center according to claim 8, characterized in that, The step of executing a control strategy matching the real-time determined chip accumulation state level includes: In response to normal cutting conditions, a silent monitoring strategy is implemented; in response to a trend of increasing chip density, a low-intensity early warning chip removal strategy is implemented; in response to a rapidly deteriorating condition, a real-time machining parameter intervention strategy is implemented; and in response to a severe chip buildup condition, a protective shutdown and powerful chip removal strategy is implemented.

10. A chip removal control system for a machining center, characterized in that, include: A processor and a memory, the memory storing computer program instructions that, when executed by the processor, implement a chip removal control method for a machining center according to any one of claims 1-9.

Images