CN111113150A - Method for monitoring state of machine tool cutter - Google Patents

Method for monitoring state of machine tool cutter Download PDF

Info

Publication number
CN111113150A
CN111113150A CN201911413104.4A CN201911413104A CN111113150A CN 111113150 A CN111113150 A CN 111113150A CN 201911413104 A CN201911413104 A CN 201911413104A CN 111113150 A CN111113150 A CN 111113150A
Authority
CN
China
Prior art keywords
current signals
machine tool
current signal
mean square
lower boundary
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN201911413104.4A
Other languages
Chinese (zh)
Other versions
CN111113150B (en
Inventor
马鹏举
童赛赛
廖志兵
崔剑
叶波
宁勇
苏庆怀
万坤
李海
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beihang University
AECC South Industry Co Ltd
Original Assignee
Beihang University
AECC South Industry Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Beihang University, AECC South Industry Co Ltd filed Critical Beihang University
Priority to CN201911413104.4A priority Critical patent/CN111113150B/en
Publication of CN111113150A publication Critical patent/CN111113150A/en
Application granted granted Critical
Publication of CN111113150B publication Critical patent/CN111113150B/en
Expired - Fee Related legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B23—MACHINE TOOLS; METAL-WORKING NOT OTHERWISE PROVIDED FOR
    • B23Q—DETAILS, COMPONENTS, OR ACCESSORIES FOR MACHINE TOOLS, e.g. ARRANGEMENTS FOR COPYING OR CONTROLLING; MACHINE TOOLS IN GENERAL CHARACTERISED BY THE CONSTRUCTION OF PARTICULAR DETAILS OR COMPONENTS; COMBINATIONS OR ASSOCIATIONS OF METAL-WORKING MACHINES, NOT DIRECTED TO A PARTICULAR RESULT
    • B23Q17/00—Arrangements for observing, indicating or measuring on machine tools
    • B23Q17/09—Arrangements for observing, indicating or measuring on machine tools for indicating or measuring cutting pressure or for determining cutting-tool condition, e.g. cutting ability, load on tool
    • B23Q17/0952—Arrangements for observing, indicating or measuring on machine tools for indicating or measuring cutting pressure or for determining cutting-tool condition, e.g. cutting ability, load on tool during machining
    • B23Q17/0957—Detection of tool breakage
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B23—MACHINE TOOLS; METAL-WORKING NOT OTHERWISE PROVIDED FOR
    • B23Q—DETAILS, COMPONENTS, OR ACCESSORIES FOR MACHINE TOOLS, e.g. ARRANGEMENTS FOR COPYING OR CONTROLLING; MACHINE TOOLS IN GENERAL CHARACTERISED BY THE CONSTRUCTION OF PARTICULAR DETAILS OR COMPONENTS; COMBINATIONS OR ASSOCIATIONS OF METAL-WORKING MACHINES, NOT DIRECTED TO A PARTICULAR RESULT
    • B23Q17/00—Arrangements for observing, indicating or measuring on machine tools
    • B23Q17/09—Arrangements for observing, indicating or measuring on machine tools for indicating or measuring cutting pressure or for determining cutting-tool condition, e.g. cutting ability, load on tool
    • B23Q17/0952—Arrangements for observing, indicating or measuring on machine tools for indicating or measuring cutting pressure or for determining cutting-tool condition, e.g. cutting ability, load on tool during machining
    • B23Q17/0961—Arrangements for observing, indicating or measuring on machine tools for indicating or measuring cutting pressure or for determining cutting-tool condition, e.g. cutting ability, load on tool during machining by measuring power, current or torque of a motor

Abstract

The invention discloses a monitoring method of machine tool cutter state, which processes the current signal of the machine tool spindle motor, obtains the upper and lower boundary curves of the current signal when the normal cutter cuts through the probability distribution theory in statistics, uses the upper and lower boundary curves as the standard for monitoring the current signal of the spindle motor, monitors the abrasion degree of the machine tool cutter through identifying the slight change of the current signal of the spindle motor, can effectively monitor the cutter state, and the upper and lower boundary curves are obtained through self-learning in the machining process of the machine tool, therefore, the monitoring method provided by the invention has the advantage of adapting to various machining conditions, in addition, the monitoring method provided by the invention monitors the cutter state through measuring the current signal, but not force signal, vibration signal and acoustic emission signal, therefore, no additional sensor, wiring and the like are needed in the machining environment of the machine tool, thereby avoiding an additional influence on the normal cutting process.

Description

Method for monitoring state of machine tool cutter
Technical Field
The invention relates to the technical field of machine manufacturing, in particular to a method for monitoring the state of a cutter of a machine tool.
Background
The cutter is an important part for numerical control machine tool machining. The high temperature and cutting force generated during the machining process make the abrasion or fracture of the tool become an inevitable problem, which is also an important and common source of problems in the numerical control machine tool machining technology. In the machining process, if rapid abrasion or breakage of the tool cannot be found in time and the machine tool stops working, the problems of reduction of machining precision, damage of the machine tool or a workpiece and the like are caused, and extra economic loss is brought. In order to avoid such a loss as much as possible, it is necessary to monitor the state of the tool during machining.
Currently, in the field of monitoring the state of a tool, most of domestic and foreign scholars monitor the state of the tool by processing, analyzing and monitoring signals of machine tool operation, such as force signals, vibration signals, acoustic emission signals, current signals and the like, to obtain characteristic information related to the state of the tool. The tool state monitoring mainly comprises four parts of signal acquisition, signal processing, feature extraction and classification or wear value estimation, wherein the classification or wear value estimation part can use methods such as data linear fitting, Gaussian fitting, neural network, convolutional neural network, Markov model, support vector machine and least square support vector machine, and the classification of the tool state is mainly some intelligent algorithms of machine learning.
In the field of tool state monitoring, machine learning samples are cutting force, current (power), acoustic emission, vibration, sound and other signals obtained by cutting a tool at each stage of wear, and these signals may change due to different cutting conditions (cutting speed, cutting depth, feed rate), may also change due to different workpiece materials to be processed, and may also change due to different machine tools. In order to ensure the accuracy of the tool state monitoring model under the specific conditions to be trained, the number of training samples for machine learning needs to be increased. Then, all training samples for different processing parameters, different workpiece materials, and different numerically controlled machine tools become very large, which requires considerable investment in manpower, material resources, and financial resources. For this reason, the tool state monitoring based on machine learning is not widely applied to actual production, and is mostly adopted in laboratories by researchers.
The existing cutter state monitoring method can not carry out self-adaptive cutter state monitoring according to different cutting conditions, and in addition, sensors, wiring and the like are additionally arranged in a machine tool machining environment for measuring force signals, vibration signals, acoustic emission signals and the like, so that the normal cutting machining is greatly influenced.
Disclosure of Invention
In view of the above, the present invention provides a method for monitoring a tool state of a machine tool, so as to provide a monitoring method capable of adaptively monitoring a tool state according to different cutting conditions.
Therefore, the invention provides a method for monitoring the state of a machine tool cutter, which comprises the following steps:
s1: acquiring a current signal of a spindle motor of a machine tool in real time, and acquiring an upper boundary curve and a lower boundary curve of the current signal according to the current signal acquired in the previous M cutting processes of a machine tool cutter; wherein M is a positive integer;
iterating through the loop through steps S2 and S3;
s2: judging whether the current signal acquired in the M +1 cutting process of the machine tool cutter exceeds an upper boundary curve and a lower boundary curve; if yes, go to step S3; if not, the process returns to step S2, if M is M + 1;
s3: judging whether the number of the parts exceeding the upper and lower boundary curves is greater than a threshold value; if yes, go to step S4; if not, the process returns to step S2, if M is M + 1;
s4: and stopping the machine tool.
In a possible implementation manner, in the monitoring method provided by the present invention, in step S1, obtaining upper and lower boundary curves of the current signal according to the current signal collected during the previous M times of cutting of the tool of the machine tool, specifically includes:
s11: dividing current signals acquired in each cutting process in the previous M cutting processes of a machine tool cutter into R sections, wherein each section of current signals is a periodic sine wave, and calculating the root mean square value of each section of current signals; wherein R is an integer greater than 1;
s12: calculating the mean value and the standard deviation of N root mean square values corresponding to each N sections of current signals according to the calculated root mean square value of each section of current signals; wherein N is an integer greater than 1;
s13: calculating an upper boundary point and a lower boundary point of each N sections of current signals according to the calculated mean value and standard deviation of N root mean square values corresponding to each N sections of current signals;
s14: and connecting the upper boundary points of the N sections of current signals, and connecting the lower boundary points of the N sections of current signals to obtain an upper and lower boundary curve.
In a possible implementation manner, in the monitoring method provided by the present invention, in step S11, the root mean square value of each segment of the current signal is calculated, and the root mean square value I of each segment of the current signal is calculated according to the following formularms：
Figure BDA0002350494610000031
Wherein, I1,I2...ItAnd t is the number of data points of each section of current signal.
In a possible implementation manner, in the monitoring method provided by the present invention, in step S12, a mean value and a standard deviation of N root mean square values corresponding to each N segments of current signals are calculated according to the calculated root mean square value of each segment of current signals, and if M is equal to 1, the mean value and the standard deviation of N root mean square values corresponding to each N segments of current signals are calculated according to the following formulas:
Figure BDA0002350494610000032
Figure BDA0002350494610000033
wherein the content of the first and second substances,
Figure BDA0002350494610000034
n root mean square values corresponding to the ith group of N current signals, i is an integer greater than 1, R is i × N, k is 1,2iIs the mean value, sigma, of N root mean square values corresponding to the ith group of N sections of current signalsiAnd the standard deviation of N root mean square values corresponding to the ith group of N sections of current signals.
In a possible implementation manner, in the monitoring method provided by the present invention, in step S12, according to the calculated root mean square value of each segment of current signal, a mean value and a standard deviation of N root mean square values corresponding to each N segments of current signal are calculated, and if M >1, the mean value and the standard deviation of N root mean square values corresponding to each N segments of current signal are calculated according to the following formulas:
Figure BDA0002350494610000041
Figure BDA0002350494610000042
wherein the content of the first and second substances,
Figure BDA0002350494610000043
n root mean square values corresponding to the ith group of N sections of current signals in the current signals collected in the nth cutting process, wherein i is an integer greater than 1, R is i multiplied by N, N is 1,2iIs the mean value, sigma, of N root mean square values corresponding to the ith group of N sections of current signalsiAnd the standard deviation of N root mean square values corresponding to the ith group of N sections of current signals.
In a possible implementation manner, in the monitoring method provided by the present invention, in step S13, the upper boundary point and the lower boundary point of each N-segment current signal are calculated according to the calculated mean value and standard deviation of N root mean square values corresponding to each N-segment current signal, and the upper boundary point and the lower boundary point of each N-segment current signal are calculated according to the following formulas:
Figure BDA0002350494610000044
Figure BDA0002350494610000045
wherein the content of the first and second substances,
Figure BDA0002350494610000046
is the upper boundary point of the ith group of N-section current signals,
Figure BDA0002350494610000047
is the lower boundary point of the ith group of N-section current signals.
In a possible implementation manner, in the monitoring method provided by the present invention, step S2, the determining whether the current signal acquired in the M +1 th cutting process of the machine tool exceeds the upper and lower boundary curves specifically includes:
s21: dividing current signals acquired in the M +1 cutting process of a machine tool cutter into R sections, wherein each section of current signal is a periodic sine wave, and calculating the root mean square value of each section of current signal; wherein R is an integer greater than 1;
s22: connecting the root mean square values of the current signals of all the sections to obtain a root mean square curve;
s23: and judging whether the root mean square curve of the current signal acquired in the M +1 cutting process of the machine tool cutter exceeds the upper and lower boundary curves.
In a possible implementation manner, in the monitoring method provided by the present invention, step S3, the determining whether the number of the portions exceeding the upper and lower boundary curves is greater than a threshold includes:
s31: and judging whether the number of the parts of the root-mean-square curve of the current signal collected within 20s, which exceed the upper and lower boundary curves, is more than 10.
The monitoring method provided by the invention can effectively monitor the state of the cutter by processing the current signal of the spindle motor of the machine tool and obtaining the upper and lower boundary curves of the current signal when the normal cutter cuts through the probability distribution theory in statistics and taking the upper and lower boundary curves as the standard for monitoring the current signal of the spindle motor and monitoring the abrasion degree of the cutter of the machine tool by identifying the slight change of the current signal of the spindle motor, and the upper and lower boundary curves are obtained by self-learning in the machining process of the machine tool, so the monitoring method provided by the invention has the advantages of adapting to various machining conditions and has important significance for realizing the online monitoring of the state of the cutter of the machine tool, in addition, the monitoring method provided by the invention monitors the state of the cutter by measuring the current signal instead of force signal, vibration signal and acoustic emission signal, therefore, additional arrangement of sensors, wiring and the like in a machining environment of the machine tool is not needed, and additional influence on normal cutting machining can be avoided.
Drawings
FIG. 1 is a flow chart of a method for monitoring the state of a tool of a machine tool according to the present invention;
FIG. 2 is a second flowchart of a method for monitoring the status of a tool of a machine tool according to the present invention;
FIG. 3 is a root-mean-square curve and upper and lower boundary curves obtained by calculating a root-mean-square value from a periodic sine wave of a current signal;
FIG. 4 is a root-mean-square curve and upper and lower boundary curves obtained by calculating a root-mean-square value from data points of fixed point numbers of a current signal;
fig. 5 is an upper and lower boundary curve obtained according to the current signals collected during the first, second, third, fourth, fifth and fifth cuts of the machine tool according to embodiment 3 of the present invention;
FIG. 6 is a root mean square curve of current signals collected during the eighth cutting of the machine tool and upper and lower boundary curves obtained from the current signals collected during the previous five cutting of the machine tool;
FIGS. 7a and 7b are views of the tool after an eighth cut;
FIG. 8 is a root mean square curve of current signals collected during the eleventh cutting of the machine tool and upper and lower boundary curves obtained from the current signals collected during the previous five cutting of the machine tool;
FIGS. 9a, 9b and 9c are views of the tool after the tenth cut;
FIG. 10 is a root mean square curve of current signals collected during the twenty-fifth cutting of the machine tool and upper and lower boundary curves obtained from the current signals collected during the previous five cutting of the machine tool;
11a, 11b and 11c are views of the tool after a twenty-fifth cut;
fig. 12 is a root-mean-square curve of current signals collected during twenty-eight times of cutting by the machine tool cutter and upper and lower boundary curves obtained from the current signals collected during the previous five times of cutting by the machine tool cutter.
Detailed Description
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, and it is obvious that the described embodiments are only illustrative and are not intended to limit the present application.
The invention provides a monitoring method of the state of a machine tool cutter, as shown in figure 1, comprising the following steps:
s1: acquiring a current signal of a spindle motor of a machine tool in real time, and acquiring an upper boundary curve and a lower boundary curve of the current signal according to the current signal acquired in the previous M cutting processes of a machine tool cutter; wherein M is a positive integer; specifically, the current data may be updated every 0.5 s;
iterating through the loop through steps S2 and S3;
s2: judging whether the current signal acquired in the M +1 cutting process of the machine tool cutter exceeds an upper boundary curve and a lower boundary curve; if yes, go to step S3; if not, the process returns to step S2, if M is M + 1;
s3: judging whether the number of the parts exceeding the upper and lower boundary curves is greater than a threshold value; if yes, go to step S4; if not, the process returns to step S2, if M is M + 1;
s4: and stopping the machine tool.
The monitoring method provided by the invention can effectively monitor the state of the cutter by processing the current signal of the spindle motor of the machine tool and obtaining the upper and lower boundary curves of the current signal when the normal cutter cuts through the probability distribution theory in statistics and taking the upper and lower boundary curves as the standard for monitoring the current signal of the spindle motor and monitoring the abrasion degree of the cutter of the machine tool by identifying the slight change of the current signal of the spindle motor, and the upper and lower boundary curves are obtained by self-learning in the machining process of the machine tool, so the monitoring method provided by the invention has the advantages of adapting to various machining conditions and has important significance for realizing the online monitoring of the state of the cutter of the machine tool, in addition, the monitoring method provided by the invention monitors the state of the cutter by measuring the current signal instead of force signal, vibration signal and acoustic emission signal, therefore, additional arrangement of sensors, wiring and the like in a machining environment of the machine tool is not needed, and additional influence on normal cutting machining can be avoided.
In the following, taking M ═ 5 as an example, the monitoring method provided by the present invention includes the following steps:
1. acquiring current signals of a spindle motor of a machine tool in real time, and acquiring upper and lower boundary curves of the current signals according to the current signals acquired in the previous five cutting processes of a machine tool cutter;
circularly iterating the following steps until the machine tool is stopped;
2. judging whether the current signal acquired in the sixth cutting process of the machine tool cutter exceeds an upper boundary curve and a lower boundary curve;
if yes, judging whether the number of the parts exceeding the upper and lower boundary curves is larger than a threshold value; if yes, stopping the machine tool; if not, judging the seventh cutting process, namely judging whether the current signal acquired in the seventh cutting process of the machine tool cutter exceeds the upper and lower boundary curves;
and if not, judging the seventh cutting process, namely judging whether the current signal acquired in the seventh cutting process of the machine tool cutter exceeds the upper and lower boundary curves.
It is to be noted that in the above monitoring method provided by the present invention, each cutting process is all cutting processes required for completing one part.
In a specific implementation, when step S1 in the monitoring method provided by the present invention is executed, the current signal of the spindle motor of the machine tool is collected in real time, and the upper and lower boundary curves of the current signal are obtained according to the current signal collected in the previous M times of cutting process of the machine tool, as shown in fig. 2, the method can be specifically implemented in the following manner:
s11: collecting current signals of a spindle motor of a machine tool in real time, dividing the current signals collected in each cutting process in the previous M cutting processes of a cutter of the machine tool into R sections, wherein each section of current signals is a periodic sine wave, and calculating the root mean square value of each section of current signals; wherein R is an integer greater than 1;
specifically, the current signal is converted into a root mean square value, so that the tiny change of the current signal along with the abrasion of the cutter can be well distinguished; in addition, the acquired current signals of the spindle motor are segmented by taking one period of a sine wave as a section and then converted into a stable root mean square value point sequence, so that the influence of large standard deviation or instability of the calculated root mean square value point sequence caused by human factors can be reduced;
s12: calculating the mean value and the standard deviation of N root mean square values corresponding to each N sections of current signals according to the calculated root mean square value of each section of current signals; wherein N is an integer greater than 1;
s13: calculating an upper boundary point and a lower boundary point of each N sections of current signals according to the calculated mean value and standard deviation of N root mean square values corresponding to each N sections of current signals;
the upper and lower boundaries of the root mean square value corresponding to the current signal of the spindle motor are calculated by utilizing the Gaussian distribution theory in statistics, the method is simple, has a mathematical theory basis and small calculation amount, can be suitable for various cutting conditions, meets the requirement of adaptivity, and can realize the online monitoring of the cutter state;
s14: and connecting the upper boundary points of the N sections of current signals, and connecting the lower boundary points of the N sections of current signals to obtain an upper and lower boundary curve.
Specifically, fig. 3 is a root-mean-square curve and upper and lower boundary curves obtained by calculating a root-mean-square value from a sine wave of one period of the current signal, fig. 4 is a root-mean-square curve and upper and lower boundary curves obtained by calculating a root-mean-square value from data points of a fixed number of points of the current signal, and the current signals used in fig. 3 and 4 are the same. Comparing fig. 3 and fig. 4, it can be seen that the upper and lower boundary curves of fig. 3 are more stable than the upper and lower boundary curves of fig. 4, which illustrates that the upper and lower boundary curves obtained by the monitoring method of the present invention by calculating a root mean square value with a periodic sine wave of a current signal are more stable, and can effectively identify the slight change of the current signal in the tool wear of the random machine tool, and the monitoring effect of the tool state of the machine tool is better.
In a specific implementation, when step S2 in the monitoring method provided by the present invention is executed, and whether the current signal collected in the M +1 th cutting process of the machine tool exceeds the upper and lower boundary curves is determined, as shown in fig. 2, the method can specifically be obtained by:
s21: dividing current signals acquired in the M +1 cutting process of a machine tool cutter into R sections, wherein each section of current signal is a periodic sine wave, and calculating the root mean square value of each section of current signal; wherein R is an integer greater than 1;
s22: connecting the root mean square values of the current signals of all the sections to obtain a root mean square curve;
s23: and judging whether the root mean square curve of the current signal acquired in the M +1 cutting process of the machine tool cutter exceeds the upper and lower boundary curves.
In a specific implementation, when step S3 in the monitoring method provided by the present invention is executed, and it is determined whether the number of the portions exceeding the upper and lower boundary curves is greater than a threshold, as shown in fig. 2, the following method may be specifically implemented:
s31: and judging whether the number of the parts of the root-mean-square curve of the current signal collected within 20s, which exceed the upper and lower boundary curves, is more than 10. Specifically, if the number of the parts of the root-mean-square curve of the current signal collected within 20s, which exceed the upper and lower boundary curves, is greater than 10, the tool is considered to be seriously worn, and at this time, the machine tool needs to be stopped and replaced. Considering the influence of random factors, it is normal that the rms curve of the current signal collected within 20s exceeds the upper and lower boundary curves only a few places.
In specific implementation, after the step S23 in the monitoring method provided by the present invention is executed, after determining whether the root mean square curve of the current signal acquired in the M +1 th cutting process of the machine tool exceeds the upper and lower boundary curves, if the root mean square curve of the current signal acquired in the M +1 th cutting process of the machine tool exceeds the upper and lower boundary curves, the step S31 is executed, and whether the number of the portions of the root mean square curve of the current signal acquired in 20S, which exceed the upper and lower boundary curves, is greater than 10; if the root-mean-square curve of the current signal collected in the M +1 th cutting process of the machine tool does not exceed the upper and lower boundary curves, then M is equal to M +1, and the process returns to step S21. After the step S31 of the monitoring method provided by the present invention is executed, whether the number of the portions of the root mean square curve of the current signal collected within 20S, which exceed the upper and lower boundary curves, is greater than 10 is judged, and if the number of the portions of the root mean square curve of the current signal collected within 20S, which exceed the upper and lower boundary curves, is greater than 10, the step S4 is executed, and the machine tool is stopped; if not, M is equal to M +1, and the process returns to step S21.
In a specific implementation, in step S1 of the monitoring method provided by the present invention, when the upper and lower boundary curves are obtained according to the current signals collected during the previous M times of cutting of the tool, M is a positive integer, that is, M may be 1, or M may also be an integer greater than 1, that is, the upper and lower boundary curves may be obtained according to the current signals collected during the first time of cutting of the tool, or the upper and lower boundary curves may also be obtained according to the current signals collected during the previous M times of cutting of the tool. Of course, the upper and lower boundary curves obtained according to the current signals collected in the previous cutting processes of the machine tool cutter are more stable than the upper and lower boundary curves obtained according to the current signals collected in the first cutting process of the machine tool cutter, and the influence of random factors can be reduced. However, it should be noted that the value of M cannot be too large, because the degree of wear of the tool increases with the increase of the number of cutting times, if the degree of wear of the tool affects the machining accuracy and damages the machine tool or the workpiece, the upper and lower boundary curves obtained according to the current signal collected during the cutting process have no reference significance, and therefore, a proper value of M needs to be selected according to different machining conditions, different machine tools, and different workpiece materials.
The following describes in detail the obtaining of the upper and lower boundary curves of the current signal in the above monitoring method provided by the present invention in two specific embodiments, where M is 1 and M > 1.
Example 1: and M is 1, namely, the upper and lower boundary curves of the current signal are obtained according to the current signal acquired in the first cutting process of the machine tool cutter.
In practical implementation, when the step S11 of the monitoring method provided by the present invention is executed to calculate the root mean square value of each current signal, the root mean square value I of each current signal can be calculated according to the following formularms：
Figure BDA0002350494610000101
Wherein, I1,I2...ItAnd t is the number of data points of each section of current signal.
In a specific implementation, when the step S12 in the monitoring method provided by the present invention is executed, and the mean value and the standard deviation of the N root mean square values corresponding to each N sections of current signals are calculated according to the calculated root mean square value of each section of current signals, the mean value and the standard deviation of the N root mean square values corresponding to each N sections of current signals may be calculated according to the following formulas:
Figure BDA0002350494610000102
Figure BDA0002350494610000103
wherein the content of the first and second substances,
Figure BDA0002350494610000104
n root mean square values corresponding to the ith group of N current signals, i is an integer greater than 1, R is i × N, k is 1,2iIs the mean value, sigma, of N root mean square values corresponding to the ith group of N sections of current signalsiAnd the standard deviation of N root mean square values corresponding to the ith group of N sections of current signals.
In a specific implementation, when step S13 in the monitoring method provided by the present invention is executed, and the upper boundary point and the lower boundary point of each N-segment current signal are calculated according to the calculated mean value and standard deviation of N root mean square values corresponding to each N-segment current signal, the upper boundary point and the lower boundary point of each N-segment current signal may be calculated according to the following formulas:
Figure BDA0002350494610000111
Figure BDA0002350494610000112
wherein the content of the first and second substances,
Figure BDA0002350494610000113
is the upper boundary point of the ith group of N-section current signals,
Figure BDA0002350494610000114
is the lower boundary point of the ith group of N-section current signals. According to the relevant statistical theory of gaussian distribution, the rms value of the current signal should be distributed with a large probability within the (μ -3 σ, μ +3 σ) interval.
And finally, executing S14, connecting the upper boundary points of the N sections of current signals, and connecting the lower boundary points of the N sections of current signals to obtain the upper and lower boundary curves of the current signals.
Example 2: and M is greater than 1, for example, M is 5, namely, the upper and lower boundary curves of the current signal are obtained according to the current signal acquired in the previous five times of cutting processes of the machine tool cutter.
In practical implementation, when the step S11 of the monitoring method provided by the present invention is executed to calculate the root mean square value of each current signal, the root mean square value I of each current signal can be calculated according to the following formularms：
Figure BDA0002350494610000115
Wherein, I1,I2...ItAnd t is the number of data points of each section of current signal.
In a specific implementation, when the step S12 in the monitoring method provided by the present invention is executed, and the mean value and the standard deviation of the N root mean square values corresponding to each N sections of current signals are calculated according to the calculated root mean square value of each section of current signals, the mean value and the standard deviation of the N root mean square values corresponding to each N sections of current signals may be calculated according to the following formulas:
Figure BDA0002350494610000121
Figure BDA0002350494610000122
wherein the content of the first and second substances,
Figure BDA0002350494610000123
n root mean square values corresponding to the ith group of N sections of current signals in the current signals collected in the nth cutting process, wherein i is an integer greater than 1, R is i multiplied by N, N is 1,2iIs the mean value, sigma, of N root mean square values corresponding to the ith group of N sections of current signalsiAnd the standard deviation of N root mean square values corresponding to the ith group of N sections of current signals.
In a specific implementation, when step S13 in the monitoring method provided by the present invention is executed, and the upper boundary point and the lower boundary point of each N-segment current signal are calculated according to the calculated mean value and standard deviation of N root mean square values corresponding to each N-segment current signal, the upper boundary point and the lower boundary point of each N-segment current signal may be calculated according to the following formulas:
Figure BDA0002350494610000124
Figure BDA0002350494610000125
wherein the content of the first and second substances,
Figure BDA0002350494610000126
is the upper boundary point of the ith group of N-section current signals,
Figure BDA0002350494610000127
is the lower boundary point of the ith group of N-section current signals. According to the relevant statistical theory of gaussian distribution, the rms value of the current signal should be distributed with a large probability within the (μ -3 σ, μ +3 σ) interval.
And finally, executing S14, connecting the upper boundary points of the N sections of current signals, and connecting the lower boundary points of the N sections of current signals to obtain the upper and lower boundary curves of the current signals.
The following describes the specific implementation of the above monitoring method provided by the present invention in detail by using a specific embodiment.
Example 3: the workpiece material to be processed is Q235, the cutting depth ap is 2mm, the cutting width ae is 10mm, the rotation speed n is 700rpm, and the feeding amount f is 100 mm/min.
FIG. 5 shows six sets of upper and lower boundary curves, which are respectively an upper and lower boundary curve (the first set of curves shown in FIG. 5) for obtaining a current signal according to a current signal collected during a first cutting process of a machine tool bit, an upper and lower boundary curve (the second set of curves shown in FIG. 5) for obtaining a current signal according to a current signal collected during a second cutting process of a machine tool bit, an upper and lower boundary curve (the third set of curves shown in FIG. 5) for obtaining a current signal according to a current signal collected during a third cutting process of a machine tool bit, an upper and lower boundary curve (the fourth set of curves shown in FIG. 5) for obtaining a current signal according to a current signal collected during a fourth cutting process of a machine tool bit, an upper and lower boundary curve (the fifth set of curves shown in FIG. 5) for obtaining a current signal according to a current signal collected during a fifth cutting process of a machine tool bit, and an upper and lower boundary curve (the upper and lower And (ii) a boundary curve (e.g., the sixth set of curves shown in fig. 5). As can be seen from fig. 5, the upper and lower boundary curves of the current signal obtained according to the current signal acquired in the previous five cutting processes of the machine tool cutter are obviously more stable than the upper and lower boundary curves of the current signal acquired according to the current signal acquired in the single cutting process of the machine tool cutter, so that the slight change of the current signal in the abrasion of the machine tool cutter can be effectively identified, and the monitoring effect of the state of the machine tool cutter is better.
The conditions of the machine tool cutter for the first seven times of cutting are that the cutting depth ap is 2mm, the cutting width ae is 10mm, the rotating speed n is 700rpm, and the feeding amount f is 100 mm/min. Before the machine tool cutter performs the eighth cutting, the cutting depth is increased by 0.45mm, namely the cutting depth ap is 2.45mm, the cutter state of the machine tool for the eighth cutting is monitored by using upper and lower boundary curves (obtained by self-learning under the condition that the cutting depth ap is 2 mm) of current signals obtained according to the current signals collected in the previous five times of cutting of the machine tool cutter, as shown in fig. 6, the root mean square curve of the current signals collected in the eighth cutting of the machine tool cutter exceeds the upper and lower boundary curves, the exceeding area is large, as shown in fig. 7a and 7b, and after the eighth cutting, the cutter is not worn seriously. Before the machine tool cutter performs the ninth cutting, the cutting depth is modified into the original cutting depth, namely the cutting depth ap is 2mm, the cutting is continued, the cutter state of the machine tool for the eleventh cutting is monitored by using upper and lower boundary curves (obtained by self-learning under the condition that the cutting depth ap is 2 mm) of current signals obtained according to the current signals collected in the previous cutting process of the machine tool cutter for the fifth cutting, as shown in fig. 8, the root mean square curve of the current signals collected in the eleventh cutting process of the machine tool cutter does not exceed the upper and lower boundary curves, as shown in fig. 9a, 9b and 9c, and after the eleventh cutting, the cutter is not worn seriously.
In summary, the root mean square curve of the current signal acquired by cutting after the cutting depth is changed exceeds the upper and lower boundary curves obtained by self-learning under the original cutting depth condition, and the root mean square curve of the current signal acquired by cutting after the cutting depth is recovered returns to the range of the upper and lower boundary curves again, which shows that the monitoring method provided by the invention is extremely sensitive to the small change of the current signal and can effectively identify the small change of the current signal, so that the wear degree of the tool bit of the machine tool is monitored by identifying the small change of the current signal of the spindle motor, and the state of the tool bit can be effectively monitored.
The machine tool continues to cut under the original machining conditions, namely, the cutting depth ap is 2mm, the cutting width ae is 10mm, the rotating speed n is 700rpm, and the feeding amount f is 100 mm/min. Monitoring the cutter state of the machine tool for performing the twenty-fifth cutting by using an upper and lower boundary curve of a current signal obtained according to the current signal acquired in the previous five times of cutting of the cutter of the machine tool, wherein as shown in fig. 10, the root mean square curve of the current signal acquired in the twenty-fifth cutting of the cutter of the machine tool already partially exceeds the upper and lower boundary curves, and as shown in fig. 11a, 11b and 11c, after the twenty-fifth cutting, the cutter really reaches the serious wear degree.
The machine tool continues to cut under the original machining conditions, namely, the cutting depth ap is 2mm, the cutting width ae is 10mm, the rotating speed n is 700rpm, and the feeding amount f is 100 mm/min. The upper and lower boundary curves of the current signal obtained according to the current signal acquired in the previous five times cutting process of the tool of the machine tool are used for monitoring the tool state of the machine tool for performing the twenty-eighth cutting, as shown in fig. 12, the situation that the root mean square curve of the current signal acquired in the twenty-fifth cutting process of the tool of the machine tool exceeds the upper and lower boundary curves is more serious.
The monitoring method provided by the invention can effectively monitor the state of the cutter by processing the current signal of the spindle motor of the machine tool and obtaining the upper and lower boundary curves of the current signal when the normal cutter cuts through the probability distribution theory in statistics and taking the upper and lower boundary curves as the standard for monitoring the current signal of the spindle motor and monitoring the abrasion degree of the cutter of the machine tool by identifying the slight change of the current signal of the spindle motor, and the upper and lower boundary curves are obtained by self-learning in the machining process of the machine tool, so the monitoring method provided by the invention has the advantages of adapting to various machining conditions and has important significance for realizing the online monitoring of the state of the cutter of the machine tool, in addition, the monitoring method provided by the invention monitors the state of the cutter by measuring the current signal instead of force signal, vibration signal and acoustic emission signal, therefore, additional arrangement of sensors, wiring and the like in a machining environment of the machine tool is not needed, and additional influence on normal cutting machining can be avoided.
It will be apparent to those skilled in the art that various changes and modifications may be made in the present invention without departing from the spirit and scope of the invention. Thus, if such modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include such modifications and variations.

Claims (8)

1. A monitoring method for the state of a machine tool cutter is characterized by comprising the following steps:
s1: acquiring a current signal of a spindle motor of a machine tool in real time, and acquiring an upper boundary curve and a lower boundary curve of the current signal according to the current signal acquired in the previous M cutting processes of a machine tool cutter; wherein M is a positive integer;
iterating through the loop through steps S2 and S3;
s2: judging whether the current signal acquired in the M +1 cutting process of the machine tool cutter exceeds an upper boundary curve and a lower boundary curve; if yes, go to step S3; if not, the process returns to step S2, if M is M + 1;
s3: judging whether the number of the parts exceeding the upper and lower boundary curves is greater than a threshold value; if yes, go to step S4; if not, the process returns to step S2, if M is M + 1;
s4: and stopping the machine tool.
2. The monitoring method according to claim 1, wherein the step S1 of obtaining upper and lower boundary curves of the current signal according to the current signal collected during the previous M times of cutting by the machine tool specifically comprises:
s11: dividing current signals acquired in each cutting process in the previous M cutting processes of a machine tool cutter into R sections, wherein each section of current signals is a periodic sine wave, and calculating the root mean square value of each section of current signals; wherein R is an integer greater than 1;
s12: calculating the mean value and the standard deviation of N root mean square values corresponding to each N sections of current signals according to the calculated root mean square value of each section of current signals; wherein N is an integer greater than 1;
s13: calculating an upper boundary point and a lower boundary point of each N sections of current signals according to the calculated mean value and standard deviation of N root mean square values corresponding to each N sections of current signals;
s14: and connecting the upper boundary points of the N sections of current signals, and connecting the lower boundary points of the N sections of current signals to obtain an upper and lower boundary curve.
3. The monitoring method of claim 2, wherein step S11 is to calculate the rms value of each current signal, and calculate the rms value I of each current signal according to the following formularms：
Figure FDA0002350494600000011
Wherein, I1,I2...ItAnd t is the number of data points of each section of current signal.
4. The monitoring method according to claim 3, wherein in step S12, the mean value and the standard deviation of the N root mean square values corresponding to each N segments of the current signals are calculated according to the calculated root mean square value of each segment of the current signals, and if M is 1, the mean value and the standard deviation of the N root mean square values corresponding to each N segments of the current signals are calculated according to the following formula:
Figure FDA0002350494600000021
Figure FDA0002350494600000022
wherein the content of the first and second substances,
Figure FDA0002350494600000023
n root mean square values corresponding to the ith group of N current signals, i is an integer greater than 1, R is i × N, k is 1,2iIs the mean value, sigma, of N root mean square values corresponding to the ith group of N sections of current signalsiAnd the standard deviation of N root mean square values corresponding to the ith group of N sections of current signals.
5. The monitoring method of claim 3, wherein in step S12, the mean and standard deviation of the N root mean square values corresponding to each N segments of the current signals are calculated according to the calculated root mean square value of each segment of the current signals, and if M >1, the mean and standard deviation of the N root mean square values corresponding to each N segments of the current signals are calculated according to the following formula:
Figure FDA0002350494600000024
Figure FDA0002350494600000025
wherein the content of the first and second substances,
Figure FDA0002350494600000026
n root mean square values corresponding to the ith group of N sections of current signals in the current signals collected in the nth cutting process, wherein i is an integer greater than 1, R is i multiplied by N, N is 1,2iIs the mean value, sigma, of N root mean square values corresponding to the ith group of N sections of current signalsiAnd the standard deviation of N root mean square values corresponding to the ith group of N sections of current signals.
6. The monitoring method according to claim 4 or 5, wherein in step S13, the upper boundary point and the lower boundary point of each N segments of current signals are calculated according to the calculated mean value and standard deviation of N root mean square values corresponding to each N segments of current signals, and the upper boundary point and the lower boundary point of each N segments of current signals are calculated according to the following formulas:
Figure FDA0002350494600000031
Figure FDA0002350494600000032
wherein the content of the first and second substances,
Figure FDA0002350494600000033
is the upper boundary point of the ith group of N-section current signals,
Figure FDA0002350494600000034
is the lower boundary point of the ith group of N-section current signals.
7. The monitoring method according to claim 1, wherein the step S2 of determining whether the current signal collected during the M +1 th cutting process of the machine tool exceeds the upper and lower boundary curves specifically comprises:
s21: dividing current signals acquired in the M +1 cutting process of a machine tool cutter into R sections, wherein each section of current signal is a periodic sine wave, and calculating the root mean square value of each section of current signal; wherein R is an integer greater than 1;
s22: connecting the root mean square values of the current signals of all the sections to obtain a root mean square curve;
s23: and judging whether the root mean square curve of the current signal acquired in the M +1 cutting process of the machine tool cutter exceeds the upper and lower boundary curves.
8. The monitoring method according to claim 7, wherein the step S3 of determining whether the number of the portions exceeding the upper and lower boundary curves is greater than a threshold value specifically includes:
s31: and judging whether the number of the parts of the root-mean-square curve of the current signal collected within 20s, which exceed the upper and lower boundary curves, is more than 10.
CN201911413104.4A 2019-12-31 2019-12-31 Method for monitoring state of machine tool cutter Expired - Fee Related CN111113150B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201911413104.4A CN111113150B (en) 2019-12-31 2019-12-31 Method for monitoring state of machine tool cutter

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201911413104.4A CN111113150B (en) 2019-12-31 2019-12-31 Method for monitoring state of machine tool cutter

Publications (2)

Publication Number Publication Date
CN111113150A true CN111113150A (en) 2020-05-08
CN111113150B CN111113150B (en) 2021-11-19

Family

ID=70506533

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201911413104.4A Expired - Fee Related CN111113150B (en) 2019-12-31 2019-12-31 Method for monitoring state of machine tool cutter

Country Status (1)

Country Link
CN (1) CN111113150B (en)

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111650417A (en) * 2020-06-05 2020-09-11 东莞市凯格精机股份有限公司 Scraper state detection method, device and system
CN111694310A (en) * 2020-06-28 2020-09-22 绍兴安迪自动化设备有限公司 Intelligent data acquisition and analysis system and method thereof
CN111774933A (en) * 2020-06-30 2020-10-16 华中科技大学无锡研究院 Method for monitoring machine tool cutter in real time and automatically replacing backup cutter and real-time monitoring system for machine tool cutter
CN111948976A (en) * 2020-07-31 2020-11-17 深圳吉兰丁智能科技有限公司 Cutter state monitoring method, non-volatile readable storage medium and electronic device
CN112008495A (en) * 2020-07-28 2020-12-01 成都飞机工业（集团）有限责任公司 Cutter damage identification method based on vibration monitoring
CN114227381A (en) * 2021-12-24 2022-03-25 浙江大学台州研究院 Machine tool cutter wear detection method

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2000105605A (en) * 1998-09-28 2000-04-11 Toyoda Mach Works Ltd Robot controller
CN1352586A (en) * 1999-05-27 2002-06-05 三洋电机株式会社 Method and apparatus for checking cutting tools
CN102073300A (en) * 2010-12-28 2011-05-25 华中科技大学 Tool wear monitoring system capable of realizing self-learning in numerical control machining state
CN103324139A (en) * 2013-06-07 2013-09-25 华中科技大学 Numerical control machine tool milling tool breakage monitoring method
CN105619178A (en) * 2016-03-09 2016-06-01 华中科技大学 Real-time detecting method of tool breakage of numerically-controlled machine tool
CN205342670U (en) * 2016-01-21 2016-06-29 上海湖心亭数控科技有限公司 Cutter current monitoring system
CN109277882A (en) * 2018-09-25 2019-01-29 江苏西格数据科技有限公司 A kind of machine tool monitoring system
CN109465676A (en) * 2018-09-13 2019-03-15 成都数之联科技有限公司 A kind of cutter life prediction technique
CN110421408A (en) * 2019-07-02 2019-11-08 绍兴安迪自动化设备有限公司 A kind of monitoring method of cutting-tool wear state

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2000105605A (en) * 1998-09-28 2000-04-11 Toyoda Mach Works Ltd Robot controller
CN1352586A (en) * 1999-05-27 2002-06-05 三洋电机株式会社 Method and apparatus for checking cutting tools
CN102073300A (en) * 2010-12-28 2011-05-25 华中科技大学 Tool wear monitoring system capable of realizing self-learning in numerical control machining state
CN103324139A (en) * 2013-06-07 2013-09-25 华中科技大学 Numerical control machine tool milling tool breakage monitoring method
CN205342670U (en) * 2016-01-21 2016-06-29 上海湖心亭数控科技有限公司 Cutter current monitoring system
CN105619178A (en) * 2016-03-09 2016-06-01 华中科技大学 Real-time detecting method of tool breakage of numerically-controlled machine tool
CN109465676A (en) * 2018-09-13 2019-03-15 成都数之联科技有限公司 A kind of cutter life prediction technique
CN109277882A (en) * 2018-09-25 2019-01-29 江苏西格数据科技有限公司 A kind of machine tool monitoring system
CN110421408A (en) * 2019-07-02 2019-11-08 绍兴安迪自动化设备有限公司 A kind of monitoring method of cutting-tool wear state

Cited By (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111650417A (en) * 2020-06-05 2020-09-11 东莞市凯格精机股份有限公司 Scraper state detection method, device and system
CN111694310A (en) * 2020-06-28 2020-09-22 绍兴安迪自动化设备有限公司 Intelligent data acquisition and analysis system and method thereof
CN111774933A (en) * 2020-06-30 2020-10-16 华中科技大学无锡研究院 Method for monitoring machine tool cutter in real time and automatically replacing backup cutter and real-time monitoring system for machine tool cutter
CN111774933B (en) * 2020-06-30 2022-06-03 华中科技大学无锡研究院 Method for monitoring machine tool cutter in real time and automatically replacing backup cutter and real-time monitoring system for machine tool cutter
CN112008495A (en) * 2020-07-28 2020-12-01 成都飞机工业（集团）有限责任公司 Cutter damage identification method based on vibration monitoring
CN111948976A (en) * 2020-07-31 2020-11-17 深圳吉兰丁智能科技有限公司 Cutter state monitoring method, non-volatile readable storage medium and electronic device
CN111948976B (en) * 2020-07-31 2022-03-15 深圳吉兰丁智能科技有限公司 Cutter state monitoring method, non-volatile readable storage medium and electronic device
CN114227381A (en) * 2021-12-24 2022-03-25 浙江大学台州研究院 Machine tool cutter wear detection method

Also Published As

Publication number Publication date
CN111113150B (en) 2021-11-19

Similar Documents

Publication Publication Date Title
CN111113150B (en) Method for monitoring state of machine tool cutter
CN109909804B (en) Tool wear damage online monitoring method based on spindle driving current and process steps
CN107738140B (en) Method and system for monitoring state of cutter and processing equipment
US20180164757A1 (en) Machine learning device, cnc device and machine learning method for detecting indication of occurrence of chatter in tool for machine tool
CN109918822B (en) Shield tunneling parameter deviation calculation method and system based on random forest regression
CN110059442B (en) Turning tool changing method based on part surface roughness and power information
CN110421408B (en) Method for monitoring wear state of cutter
Bombiński et al. Sensor signal segmentation for tool condition monitoring
Jemielniak Contemporary challenges in tool condition monitoring
CN108445838A (en) A kind of numerically-controlled machine tool processing quality analysis method, grader and equipment
US20230004152A1 (en) Method for monitoring and/or predecting machining processes and/or machnining outcomes
Bhattacharyya et al. On-line tool condition monitoring in face milling using current and power signals
Yan et al. A hybrid method for on-line performance assessment and life prediction in drilling operations
CN112207631B (en) Method for generating tool detection model, method, system, device and medium for detecting tool detection model
CN116021339B (en) Method and related device for monitoring cutting force of main shaft of numerical control machine tool
Oberle et al. A use case to implement machine learning for life time prediction of manufacturing tools
CN103500251A (en) Optimized machining method of tool cutting amount in numerical control milling
CN115922442A (en) Cutter grinding damage real-time monitoring method based on spindle vibration signal and related device
CN114986258A (en) Cutter state monitoring method and device, computer equipment and system
CN114800040A (en) Cutter wear monitoring method and system based on process-state data correlation
CN116690313B (en) Failure monitoring method for machining cutter of web plate of aircraft structural member
CN104503361A (en) Multimodal fusion based gear machining process tool change decision method
CN113570254A (en) Industrial data quality analysis method
CN117331344A (en) Multi-signal quality monitoring and controlling system in numerical control machining process
CN109396956B (en) Intelligent monitoring method for hob state of numerical control gear hobbing machine

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant
CF01 Termination of patent right due to non-payment of annual fee
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20211119