JP6575814B2 - Process state monitoring method and system for work machine - Google Patents

Description

  The present invention relates to a processing state monitoring method and a system for a working machine for monitoring a processing state by the processing tool when the processing tool or workpiece is rotated to perform processing on the workpiece through the processing tool. .

  In general, various machine tools are used to process a workpiece through a processing tool. For example, in the boring process, a boring tool provided with a boring tool (cutting edge) is attached to a rotating spindle (spindle) of a machine tool, and the boring tool is sequentially fed along a pilot hole while rotating at a high speed. A highly accurate hole is processed at a predetermined position with the cutting edge diameter.

  In this type of work machine, bending due to cutting resistance is likely to occur in the spindle, processing tool, and workpiece. Then, due to this bending, vibration is induced in the machining tool or workpiece, and this vibration may become chatter (including so-called regenerative chatter) and appear in machining.

  Therefore, for example, Patent Document 1 proposes a vibration determination method and a vibration determination apparatus that can accurately determine the type of chatter vibration. In this vibration discriminating method, the first step of detecting the vibration in the time domain of the rotating rotating shaft and the rotation speed of the rotating shaft, and the vibration in the frequency domain are calculated based on the vibration in the time domain. A second step of acquiring a plurality of peak values in the frequency domain vibration and a peak frequency taking the peak value, and a rotation-synchronized vibration determination range for determining the type of chatter vibration for each peak value. The type of chatter vibration based on the relationship between the third step determined using the peak frequency and the rotation frequency determination range determined using the peak frequency and the peak frequency for each peak value. And a fourth step of discriminating the above.

  JP 2013-837 A

  However, in Patent Document 1 described above, the type of chatter vibration is determined after chatter actually occurs. Therefore, since the processing for the generated chatter vibration is performed, the influence of the chatter vibration is generated on the work member, and there is a possibility that a highly accurate machining process may not be performed.

  The present invention solves this type of problem, and with a simple process and configuration, it is possible to perform a good response before the occurrence of chatter, and to prevent the occurrence of chatter as much as possible. It is an object of the present invention to provide a machining state monitoring method and system for a work machine capable of efficiently performing accurate machining operations.

  The present invention relates to a processing state monitoring method and a system for a working machine for monitoring a processing state by the processing tool when the processing tool or workpiece is rotated to perform processing on the workpiece through the processing tool. Is.

  This machining state monitoring method is a frequency spectrum comprising a step of inputting the rotation number, the number of blades and the natural frequency of a machining tool, and a frequency and acceleration obtained by analyzing the vibration of the machining tool or workpiece during machining by Fourier series expansion. And a TPF peak that is a peak acceleration at the tool passing frequency calculated from the number of rotations and the number of blades is compared with a TPF threshold that is a peak threshold of the tool passing frequency preset in the frequency spectrum. The number of times that the TPF peak has exceeded the TPF threshold is displayed on the TPF threshold value exceeding integrated display section, and the f0 peak that is the peak acceleration at the natural frequency is preset in the frequency spectrum. The number of times that the f0 peak has exceeded the f0 threshold is compared with the f0 threshold that is the peak threshold of the natural frequency. A step of displaying the example integrated display unit has a step of displaying the change display unit compares the number of times exceeding the number and f0 threshold exceeded TPF threshold, the.

  Further, in this machining state monitoring method, the step of setting the vibration at the time of idling of the machine spindle as a threshold, the step of determining whether or not the machining vibration detected during machining of the machine spindle exceeds the threshold, And a step of analyzing the vibration of the machining tool or the workpiece by Fourier series expansion when it is determined that the machining vibration exceeds the threshold value.

  Further, the machining state monitoring system includes an input setting unit for inputting the rotation number, the number of blades, and the natural frequency of the machining tool, and the frequency and acceleration obtained by analyzing the vibration of the machining tool or the workpiece during machining by Fourier series expansion. A frequency spectrum display unit for displaying a frequency spectrum comprising: a TPF peak that is a peak acceleration at a tool passing frequency calculated from the rotation speed and the number of blades, and a peak of a tool passing frequency preset in the frequency spectrum. Compared with the TPF threshold value, which is a threshold value, a TPF threshold value exceeding display section for displaying the number of times the TPF peak has exceeded the TPF threshold value, and an f0 peak value representing the peak acceleration at the natural frequency are preliminarily displayed in the frequency spectrum It is compared with the f0 threshold that is the peak threshold of the set natural frequency, and the f0 peak is the f0 threshold. Includes a f0 threshold beyond integration display section for displaying the number of times exceeds a change display unit for displaying by comparing the number of times exceeding the number and f0 threshold exceeded TPF threshold, the.

  In the machining state monitoring method and system according to the present invention, when a workpiece is machined through a machining tool, the correlation between the peak occurrence frequency at the tool passing frequency (TPF) and the peak occurrence frequency at the natural frequency is obtained. Can see. From the change ratio of the relationship between the two, for example, it is possible to quickly and accurately determine that the workpiece processing is in a predictive period in which the workpiece is shifted from the normal state to chatter vibration (abnormal state). For this reason, it can respond quickly and effectively before chatter vibration occurs.

  1 is a schematic explanatory diagram of a work machine to which a work machine monitoring state monitoring system according to an embodiment of the present invention is applied.   It is explanatory drawing of the controller which comprises the said process state monitoring system.   It is a structure explanatory drawing of the display unit which comprises the said process state monitoring system.   It is explanatory drawing which shows the change of the process state of a workpiece | work by the relationship between time and a vibration.   It is explanatory drawing of the spectrum immediately after the process start of the said workpiece | work.   It is explanatory drawing of the spectrum of the process initial stage of the said workpiece | work.   It is explanatory drawing of the spectrum of a chatter precursor period.   It is explanatory drawing of the spectrum of the state which the said chattering sign period advanced.   It is explanatory drawing of the spectrum at the time of chatter.

  As shown in FIG. 1, a working machine monitoring state monitoring system 10 according to an embodiment of the present invention is applied to a machine tool 12.

  The machine tool 12 includes a spindle (main shaft) 18 that is rotatably provided in a housing 14 via a bearing 16, and a boring bar (processing tool) 20 that is detachably attached to the spindle 18. A boring bar 22 is attached to the tip of the boring bar 20. A work W is placed on the work table 24.

  The machining state monitoring system 10 includes an acceleration sensor 26 attached to the side of the housing 14 or a microphone 28 that obtains vibration sound using sound waves in order to detect vibration generated when machining by the boring bar 20 is started. At least one is provided. The acceleration sensor 26 and / or the microphone 28 are connected to a controller 30, and the controller 30 is connected to a machine control panel 32. The machine control panel 32 controls the machine tool 12 and is connected to the control operation panel 34.

  As shown in FIG. 2, the controller 30 includes an arithmetic unit (arithmetic mechanism) 38 that amplifies and takes in mechanical vibration (machining vibration) detected by the acceleration sensor 26 and / or the microphone 28 by an amplifier and filter circuit 36. Prepare.

  An input setting unit (input setting unit) 40 for inputting the number of rotations of the spindle 18, the number of blades of the cutting tool 22, the natural frequency, and the like is connected to the arithmetic unit 38. In the input setting unit 40, values such as a predictive threshold value and a warning threshold value for monitoring and identification determination can be set.

  Connected to the arithmetic unit 38 are a machining state determination unit 44 and an output unit 46 for outputting a spindle speed adjusted by arithmetic processing described later and a count-up signal exceeding a threshold value. The calculation unit 38 is connected to a display unit 48 that displays calculation results and detection results on the screen.

  As shown in FIG. 3, the display unit 48 includes a first total power display unit 50a, a second total power display unit 50b, a frequency spectrum display unit 52, a frequency flow display unit 54, an over-threshold integration display unit 56, and a change display unit. 58. These are displayed on a single display screen, and a detailed description of each will be described later.

  A machining state monitoring method by the machining state monitoring system 10 configured as described above will be described below.

  As shown in FIG. 1, in the machine tool 12, the spindle 18 to which the boring bar 20 is attached is driven to rotate along the prepared hole Wa of the workpiece W. Then, the boring bar 20 moves relatively to the prepared hole Wa side of the workpiece W. For this reason, the boring bar 20 rotates, and boring is performed on the inner wall surface constituting the prepared hole Wa via the cutting tool 22 attached to the boring bar 20.

In the controller 30, before starting machining, the vibration at the time of idling of the spindle 18 is acquired by the acceleration sensor 26 and / or the microphone 28, and this value is set as an allowable value (threshold value). Then, when the machining is started by the spindle 18 and the machining vibration taken in via the amplifier and filter circuit 36 exceeds an allowable value, arithmetic analysis by Fourier transform (Fourier series expansion) is performed.
Specifically, the temporal vibration f (t) is

Is the cosine harmonic component Fourier coefficient of frequency J, and b _j is the sine harmonic component Fourier coefficient of frequency J.

The Fourier coefficient for frequency J is a _j = 1 / 2T∫f (t) cos (
Then, Fourier series expansion is performed. The integration interval is 0 to T, and this integration interval T is an integral multiple of the period 1 / J. Here, it is limited to the frequency at which chatter actually occurs, for example, 10 Hz to 10,000 Hz.

Next, in the display unit 48, the total power indicated by the sum of the values (G ² ) obtained by squaring the acceleration (G) of the machining vibration is displayed on the first total power display unit 50a and the second total power display unit 50b. In the first total power display section 50a, the sum of values obtained by squaring acceleration is displayed on the vertical axis, and the elapsed time in seconds is displayed on the horizontal axis (short-time display). In the second total power display unit 50b, the sum of values obtained by squaring acceleration is displayed on the vertical axis, and the elapsed time in minutes is displayed on the horizontal axis (long-time display).

  Here, for the vibration of the total power, a two-stage setting of a predictive threshold value and a warning threshold value set separately is possible. Further, when machining vibration exceeding a threshold value occurs during machining, the signal is selectively output to the outside through the output unit 46.

  The display unit 48 is provided with a frequency spectrum display unit 52. The frequency spectrum display unit 52 displays a frequency spectrum having the frequency Hz calculated by Fourier analysis on the horizontal axis and the acceleration G on the vertical axis. Also in this frequency spectrum display unit 52, two threshold values are set: a predictive threshold value for acquiring change data corresponding to the peak frequency of machining vibration, and a warning threshold value as an alarm when an abnormal frequency peak appears. Is done.

  When the acquired machining vibration exceeds the sign threshold value set in the frequency spectrum display unit 52, the count-up signal is sent to the threshold value excess integration display unit 56 and displayed on the laminated graph. Here, it is determined whether the counted up signal is a tool passing frequency (Tool-Passing-Frequency) (hereinafter also referred to as TPF), a natural frequency (hereinafter also referred to as f0), or any other signal. These are displayed as a laminated graph in a predetermined TPF display field (TPF threshold value exceeding display part) 56a, natural frequency display field (f0 threshold value exceeding display part) 56b, and other display field 56c. There is also a frequency flow display 54 that shows how the strongest frequency appears over time.

  Hereinafter, TPF, natural frequency, and other frequency peaks will be described. FIG. 4 shows machining vibrations when boring with a multi-blade. The vertical axis is the acceleration G, which is the magnitude of the vibration amount, and the horizontal axis is time.

  In FIG. 4, the left side is a machining start point, and represents a machining state without chattering until the middle of machining. Then, it is recognized that the chatter state is reached when the acceleration G exceeds ± 1 G during the machining.

  Here, as parameters for determining chatter vibration, there are a tool passing frequency at which the cutting edge comes into contact with the workpiece W = (number of rotations of the spindle / 60) × number of teeth and natural frequency. The relationship between the two parameters TPF, the output of f0, and chatter vibration will be described with reference to FIGS. 5 to 9 are explanatory diagrams of frequency spectra obtained by Fourier transforming vibrations. The vertical axis represents vibration acceleration G, and the horizontal axis represents frequency Hz.

  FIG. 5 shows a frequency spectrum for a while after the start of machining in the boring process shown in FIG. This is a state without chatter, and the spectrum has a peak acceleration at 67 Hz which is the frequency of TPF. The acceleration of this vibration is as small as 0.015 G, and normal cutting without chatter is performed.

  Furthermore, when the processing time has elapsed, as shown in FIG. 6, many other peak frequencies other than TPF begin to occur. Since peak acceleration appears at other frequencies, the TPF signal is unclear. The peak acceleration of TPF is 0.013G which is substantially the same as in FIG.

  As shown in FIG. 7, as the chatter progresses, the amount of vibration (peak acceleration) other than TPF further increases and the clarity of the TPF peak acceleration decreases as chatter begins. In this state, many vibration peaks appear in addition to the vibration peak of TPF, and the presence or absence of the peak vibration of the natural frequency is unclear.

  Then, as shown in FIG. 8, when machining in the predictive period proceeds, the peak acceleration of 323 Hz, which is the natural frequency, becomes conspicuous. This peak acceleration is as low as about 0.04 G, is a sign of chatter, and is not considered to be a complete chatter state. In the TPF, as in the state of FIG. 7, the peak acceleration is maintained at about 0.013 G, but it is difficult to discriminate because the vibration peak of the natural frequency is abruptly exceeding 0.04 G.

  When the processing time further advances and the complete chatter phase (regenerative chatter) starts, as shown in FIG. 9, the peak acceleration of the natural frequency (323 Hz) becomes a value that is seven times larger than the state shown in FIG. Become. At that time, vibration of TPF is also generated, but since chatter vibration is greatly generated, it cannot be discriminated.

  As described above, in the state where there is no chatter, the vibration of the TPF appears greatly as the cutting vibration, while in the state where chatter occurs, the vibration of the natural frequency appears larger than the vibration of the TPF. For this reason, there are two parameters related to chatter: TPF and natural frequency. By monitoring the transition of peak acceleration of TPF and natural frequency, the predictive period of chatter can be determined.

  Therefore, in the present embodiment, as a preparation in advance, the natural frequency that is one of the monitoring parameters is input to the input setting unit 40. The natural frequency can be detected by an impact excitation method using an FFT analyzer. It can also be calculated from the vibration frequency of the already acquired vibration. The TPF, which is one of the monitoring parameters, is calculated from the number of rotations of the spindle 18 and the number of blades of the cutting tool 22, so the rotation number and the number of blades are input to the input setting unit 40.

  Next, as described above, the spindle 18 rotates to process the workpiece W, and a frequency spectrum composed of a frequency and an acceleration obtained by analyzing the processing vibration by Fourier series expansion is obtained. Then, the peak acceleration (TPF peak) at the tool passing frequency is compared with the tool passing frequency peak threshold (TPF threshold) preset in the frequency spectrum. The number of times that the TPF peak has exceeded the TPF threshold value is displayed in the TPF display field 56 a of the threshold value excess display section 56.

  On the other hand, the peak acceleration (f0 peak) at the natural frequency is compared with the peak threshold (f0 threshold) of the natural frequency preset in the frequency spectrum. The number of times that the f0 peak has exceeded the f0 threshold value is displayed in the natural frequency display field 56b of the threshold value exceeding integrated display section 56.

  Further, the change display unit 58 displays the number of times the TPF threshold has been exceeded and the number of times the f0 threshold has been exceeded in comparison. Specifically, the change ratio of the number of times of exceeding the f0 threshold value / the number of times of exceeding the TPF threshold value is instantaneously determined in order of time and displayed on a chart. Parameters to be compared in this chart can be set on a separate screen including gain selection that can weight the integrated values of the number of times exceeding the threshold. When the change ratio displayed on the change display unit 58 exceeds a predetermined value, it can be determined that the chatter state is present. Further, when the change ratio exceeds a certain value, it is also possible to immediately apply feedback for avoiding chatter by changing the rotation speed or changing the feed speed.

  In this case, in this embodiment, when the workpiece W is processed through the boring bar 20, the correlation between the number of peak occurrences at the tool passing frequency (TPF) and the number of peak occurrences at the natural frequency is observed. Can do. From the change ratio of the relationship between the two, for example, it is possible to quickly and accurately determine that the workpiece processing is in a predictive period in which the workpiece is shifted from the normal state to chatter vibration (abnormal state). For this reason, it can respond quickly and effectively before chatter vibration occurs.

  In addition, the change display unit 58 displays the number of times the TPF threshold has been exceeded and the number of times the f0 threshold has been exceeded in comparison. Therefore, the machining state can be visually confirmed in real time, and it becomes a criterion for judging the accuracy of judgment criteria, error judgment, and the like, and the workability can be easily improved.

DESCRIPTION OF SYMBOLS 10 ... Machining state monitoring system 12 ... Machine tool 14 ... Housing 18 ... Spindle 20 ... Boring bar 22 ... Byte 26 ... Accelerometer 28 ... Microphone 30 ... Controller 32 ... Machine control board 34 ... Control operation board 38 ... Arithmetic unit 40 ... Input Setting unit 44 ... Processing state determination unit 46 ... Output unit 48 ... Display units 50a, 50b ... Total power display unit 52 ... Frequency spectrum display unit 54 ... Frequency flow display unit 56 ... Over-threshold integration display unit 58 ... Change display unit

Claims (3)

A processing state monitoring method for a working machine for monitoring a processing state by the processing tool when the processing tool or the workpiece rotates to perform processing on the workpiece through the processing tool,
Inputting the number of rotations, the number of blades and the natural frequency of the processing tool;
Obtaining a frequency spectrum composed of a frequency and acceleration obtained by analyzing the vibration of the machining tool or the workpiece during machining by Fourier series expansion;
A TPF peak that is a peak acceleration at a tool passing frequency calculated from the number of rotations and the number of blades is compared with a TPF threshold that is a peak threshold of the tool passing frequency preset in the frequency spectrum, and the TPF peak Displaying the number of times the TPF threshold value is exceeded on the TPF threshold value exceeding integration display unit;
The f0 peak that is the peak acceleration at the natural frequency is compared with the f0 threshold that is the peak threshold of the natural frequency preset in the frequency spectrum, and the number of times the f0 peak exceeds the f0 threshold is calculated as follows: a step of displaying on the f0 threshold value exceeding display unit;
Comparing the number of times the TPF threshold has been exceeded with the number of times the f0 threshold has been exceeded, and causing the change display unit to display the result,
A processing state monitoring method for a work machine, comprising: The machining state monitoring method according to claim 1, wherein a step of setting a vibration during idling of the machine spindle as a threshold value;
Determining whether machining vibration detected during machining of the machine spindle exceeds the threshold;
Analyzing the vibration of the processing tool or the workpiece by the Fourier series expansion when it is determined that the processing vibration exceeds the threshold;
A processing state monitoring method for a work machine, comprising: A working state monitoring system for a work machine for monitoring a working state by the working tool when the working tool or the work rotates to perform the work processing on the work through the working tool;
An input setting unit for inputting the number of rotations, the number of blades and the natural frequency of the processing tool;
A frequency spectrum display unit for displaying a frequency spectrum composed of a frequency and acceleration obtained by analyzing the vibration of the machining tool or the workpiece at the time of machining by Fourier series expansion;
A TPF peak that is a peak acceleration at the tool passing frequency calculated from the number of rotations and the number of blades is compared with a TPF threshold that is a peak threshold of the tool passing frequency preset in the frequency spectrum, and the TPF A TPF threshold value overrun display section for displaying the number of times the peak has exceeded the TPF threshold value;
The f0 peak that is the peak acceleration at the natural frequency is compared with the f0 threshold that is the peak threshold of the natural frequency preset in the frequency spectrum, and the number of times that the f0 peak exceeds the f0 threshold is displayed. An f0 threshold value overrun display section to be made;
A change display section for comparing and displaying the number of times the TPF threshold is exceeded and the number of times the f0 threshold is exceeded;
A working state monitoring system for a work machine, comprising: