JPH0332553A - Tool breakage detecting device - Google Patents

Abstract

PURPOSE:To detect the breakage of a tool easily and positively by providing a circuit means, performing data processing on the basis of detection signals from AE sensors, with a fuzzy control part for performing the fuzzy inference of tool breakage using the data output. CONSTITUTION:Pseudo AE signals containing the frequency components of AE signals generated at the time of tool breakage are generated from a pseudo AE signal generator 34 and received at both AE sensors 8, 10 to be set in their receiving levels. The breakage of a drill 4 is then detected at a circuit means on the basis of the detection signals of both AE sensors 8, 10. In this case, this circuit means performs the fuzzy inference of the source of the AE signals at its fuzzy control part 28 on the basis of the frequency components and rising characteristics of the respective AE signals of both AE sensors 8, 10 using AE signals related to the tool breakage, the wave receiving time difference between the respective AE signals and the difference of distance from the tool to the respective AE sensors 8, 10. The estimating accuracy of the source, which is the source of the tool breakage, is thereby heightened so as to detect the breakage of the drill 4 more accurately.

Description

DETAILED DESCRIPTION OF THE INVENTION (Industrial Application Field) The present invention provides a machine tool that is attached to a machine tool and driven to rotate.
The present invention relates to a tool breakage detection device for a machine tool that detects breakage of a cutting tool such as a drill, a cutting tool, a reamer, etc., and particularly relates to detecting the breakage using fuzzy inference.

(Prior art) When a machine tool, such as a drill, is cutting a workpiece, the machine tool generates an ultrasonic frequency vibration called an AE multiplied signal (acoustic emission signal), but when the drill breaks, A conventional technique has already been proposed in which the breakage of a tool such as a drill is detected by utilizing the sudden rise in the amplitude of the AE multiplied signal.

The one described in Japanese Patent Application Laid-Open No. 61-82161 is one of the prior art techniques. In the prior art described in this publication, a pseudo AE signal generator capable of generating a pseudo AE multiplied signal including a frequency component at the time of breakage is provided at the tool position of the workpiece before machining work, and this pseudo AE signal generator The pseudo AE signal generated from the AE signal generator is received by the AE sensor installed on the table when the workpiece is mounted, and the receiving level of the AE sensor is set. Tool breakage is detected based on the logical product of the AE multiplied signal frequency component and its rise characteristic.

However, in this conventional technology, the identification of the frequency component of the AE multiplied signal at the time of tool breakage and the detection of the rise characteristic were performed by simply comparing binary levels or comparing with a reference value, respectively. For the AE multiplied signal, AE caused by a cause different from this tool breakage.
Highly reliable tool breakage detection is performed by accurately separating multiple signals, such as signals generated from chips on the workpiece, electrical noise associated with the opening and closing of a solenoid, and impact sound when an object contacts the workpiece. I couldn't do that.

In contrast to this conventional technique, a conventional technique described in Japanese Unexamined Patent Publication No. 117449/1983 has been proposed.

In this conventional technology, a plurality of AE sensors are provided,
By estimating the source of the AE multiplied signal from the E multiplied signal detection time difference and comparing the estimation with the actual tool position,
The AE multiplied signal due to tool breakage and the AE multiplied signal due to other causes can be distinguished, thereby increasing the reliability of tool breakage detection.

(Problem to be Solved by the Invention) However, in this prior art, as is clear from the publication, it is necessary to change the program depending on the position when installing the AE sensor, and also to identify the source of the AE signal. The problem was that it required complex calculations.

The present invention has been made in view of these problems, and it is possible to reliably distinguish between the AE multiplied signal due to tool breakage and the AE multiplied signal obtained due to other causes, and the installation of the AE sensor is simplified. It is an object of the present invention to provide a highly reliable tool breakage detection device that can easily and reliably detect tool breakage without being affected by the type of tool or cutting conditions due to differences in position.

(Means for Achieving the Problem) In order to achieve such an object, the tool breakage detection device of the present invention detects A that occurs when a tool such as a drill breaks.
pseudo AE multiplied signal generation means including an E multiplied signal frequency component;
at least two AE sensors that are installed near the workpiece in a machine tool body that machine the workpiece and output detection signals corresponding to the AE multiplied signal detection; and based on the detection signals from both of the AE sensors. the frequency component of the AE signal,
circuit means for data processing the rise characteristics, the AE multiplied signal reception time difference in each AE sensor, and the distance difference from the tool to each AE sensor, and the circuit means uses the data output thereof to process the data. It is characterized by being equipped with a fuzzy control section that makes fuzzy inferences about tool breakage.

(Function) Before machining, the pseudo AE signal generation means generates a pseudo AE including the AE multiplied signal frequency component that occurs when a tool such as a drill breaks.
By generating the double signal and receiving the generated pseudo AE double signal with both AE sensors, the reception levels of the two AE sensors can be set. Then, the circuit means detects tool breakage based on the detection signals from both AE sensors. In this case, this circuit means has a fuzzy control unit provided therein to adjust the A of each of the two AE sensors.
AE multiplied signal related to tool breakage with frequency components and rise characteristics; AE multiplied signal whose source is tool breakage; and AE multiplied signal based on tool breakage, although there is no AE multiplied signal. Double SignalAlthough it is possible to detect the AE double signal that is generated from other sources, the source of the AE double signal can be detected by using the difference in reception time of the AE double signal at each AE sensor and the difference in distance from the tool to each AE sensor. is estimated by fuzzy inference, the accuracy of estimating the source of the tool breakage increases, and as a result, the breakage of the tool can be detected more accurately. As a result, tool breakage can be detected without the need for complicated calculation programs for identifying the source of the AE multiplied signal.

(Example) Hereinafter, an example of the present invention will be described in detail with reference to the drawings.

FIG. 1 is a block diagram of a tool breakage detection device according to an embodiment of the present invention, and FIG. 2 shows A obtained from the bed of a drill press as a machine tool body during cutting with a drill,
Fig. 3 is an amplitude waveform diagram of the E signal, and Fig. 3 is a frequency band diagram showing the <AE multiplied signal frequency band based on the Fr loss of the drill and the <AE multiplied signal frequency band based on normal cutting processing. . In Fig. 2, the period T1 is the AE multiplied signal amplitude during normal cutting, and the period T2 is the A due to the breakage of the drill.
The E times signal amplitude is shown respectively. In addition, in Fig. 3, a indicates the AE multiplied signal when the drill is broken, b indicates the AE multiplied signal during normal cutting, and the frequency band of the maximum power of the AE multiplied signal due to the broken drill is around 300 kHz. The frequency band of the maximum power of the doubled signal is 50kl (around z).

In Fig. 1, 2 is a workpiece placed on the bed (not shown) of a drilling machine as the machine tool body, 4 is a drill as a tool located above the machining position of this workpiece 2, and 6 is a drill. 4, the drill control unit controls the vertical and horizontal movement, rotation speed, etc., and 8.10 each controls the workpiece 2.
The AE sensor is attached to a predetermined position above the sensor and detects the received AE multiplied signal, and outputs a detection output at a frequency corresponding to the AE multiplied signal. In this case, for convenience of explanation, the AE sensor 8. The detection output of each IO will be referred to as the AE multiplied signal.

12a and 12b are amplifiers that amplify the respective outputs of the AE sensor 6.8, and 14a and 14b are amplifiers! In each AE multiplied signal amplified by 2a and 12b, the AE multiplied signal related to the breakage of the drill 4 is 300 kIz/i!
! Wavenumber band fraction is Xl! 3 with the purpose of making people feel welcome
The 00kHz bandpass filters 16a and 16b are 50kI-bandpass filters intended to pass the 50kIz frequency band component of the AE multiplied signal during normal cutting with the drill 4 in each AE multiplied signal amplified by the amplifiers 12a and 12b, respectively.
1z band pass filter, 18a-18d are each a detector, 20a and 20b are each a detector [8a, L
Differentiate each output of 8c and output a differentiated output that rises rapidly for signals A and E in period T2 in order to distinguish between normal cutting during period TI in Fig. 2 and breakage of the drill 4 during period T2. The differentiating circuits 22a and 22b receive the outputs of the differentiating circuits 20a and 20b, respectively, and receive the 300 kI waves received by each AE sensor 6.8.
−(AE by comparing and detecting the AE times signal reception time of z)
Comparator for obtaining data for determining the source of the AE signal associated with tool breakage for sensor 6.8; 2;
4 is the reception time t! of the AE multiplied signal received by the AE sensor 8 based on the respective outputs of the comparison 2m22a and 22b.
and the AE multiplied signal reception time t2 received by the AE sensor 10
This is a termer that outputs data related to the AE multiplied signal generation source by outputting the difference (AE signal reception time difference t!-t2). If you keep this timer 24, can you compare it? 3
22a starts the timer operation, and the comparator 22b starts the timer operation.
The timer operation is stopped at the output of , and the time difference between the start and stop is defined as the above-mentioned time difference tl-t2. 26 specifies the position of the drill 4 when it moves over the workpiece 2 under the control of the drill control unit 6, and thereby estimates the source of the A and E signals for the AE sensor 6.8. To output the data for the drill 4 to A
This is a calculation unit that calculates and outputs the distance difference dl-d2 based on each data of the distance @dl to the E-sensor 6 and the distance d2 from the drill 4 to the AE sensor 8.

28 is the output ■11 of the detector 1.8a, the output VI2 of the detector I8b, the output VI3 of the differentiating circuit 20a, and the detector 18c.
Output V21. Based on the breakage of the drill 4 at the output V22 of Bl8d, the output V23 of the differential circuit 20b, etc.
In addition to the fuzzy inference for distinguishing the AE multiplied signal based on the cause of the AE multiplied signal, the output T of the timer 24 and the output of the arithmetic unit 26 are used to calculate A fuzzy control unit comprehensively infers the breakage of the drill 4 by also performing fuzzy inference for estimating the source, and 30 is a breakage output S1 from the fuzzy control unit 28.
An output circuit 32 outputs the abnormal output S2 from the fuzzy control section 28.

The fuzzy control unit 28 uses the fuzzy labels ■ to ■ shown below and the fuzzy labels shown in FIGS. The membership function of the condition part shown in FIG. 5(e) and the membership function of the conclusion part shown in FIG. 5 are stored. These membership functions are used to apply pseudo AE to workpiece 2 before machining work.
A signal generator 34 is mounted, and this pseudo AE signal generator 34
By driving through the drive circuit 38 with the value set by the level setting device 36, an AE multiplied signal and a pseudo AE multiplied signal of the same power spectrum as when the drill 4 breaks and during cutting are generated. can be easily created.

Here, the fuzzy rules stored in the fuzzy control unit 28 are shown below.

■ if Vll= PS, Vt2= PL, Vt3=
PL, V21=PS.

V22= PL, V23= PL, T= PL, L=
NLthen breakage ■ if Vt1= PS, Vt2= PL, Vt3=
PL, V21=PS.

V22= PL, V23= PL T= NL, L=
PLthen breakage ■ if Vll-PS, Vt2=PL, Vt3=PL
J21=PS.

Vt2=PL, Vt3=PL, T=ZR, L=zR
Then breakage ■ HVt1= PL, Vt2= PL, Vt3= P
L, V21 = PL.

V22= PL, V23= PL, T= PL, L
= NLthen abnormality■ HVt1= PL, Vt2= PL, Vt3= P
L V21=PLV22=PL, Vt3=PL,
T=NL, L=PLthen abnormality■ H'V11=PL, V12=PL, V13=PL,
Vt1=PL.

Vt2=PL, Vt3=PL, T=ZR, L=ZRt
hen abnormality Here, ir is the conditional part, t h e n is the conclusion part, and V 11 , V21i;! 300kHz band AE
Double signal V12. V22 is the AE multiplied signal V in the 50kHz band
I3 and V23 are AE multiplied signal differential outputs in the 300 kHz band, T is the timer 24 output, and L is the $126 output. Further, NL, . . . PL are fuzzy label names to which each variable belongs.

To explain the operation, the fuzzy control section 28 includes each of the detectors 18a to 18d and a differentiating circuit 20a.

20b, timer 24, and arithmetic unit 26, respectively, variables V l 1 to V f 3. V21-V23
Based on T and L, membership values that match the membership functions corresponding to each fuzzy rule are determined from FIGS. 4(a) to 4(e), respectively. Then, for each fuzzy rule, select the smaller membership value of the variable of each condition part (MIN calculation), and use this selected membership value to determine each fuzzy rule's breakage or abnormality of the drill 4 from FIG. The membership functions are cut and the membership functions related to all the cut fuzzy rules are superimposed (MAX calculation) to obtain the final superposed membership function regarding breakage or abnormality of the drill 4.

Data Sl and S2 regarding the breakage or abnormality of the drill 4 determined by determining, for example, the center of gravity of this superposition membership function are outputted to the corresponding output circuits 30 and 32, respectively.

Note that a detailed explanation of the application of the above fuzzy rules is well known and will therefore be omitted.

(Effects of the Invention) As is clear from the above explanation, according to the present invention, the value of the identified AE multiplied signal frequency component, the detected value of the rapidly rising signal, and the AE multiplied signal received by each AE sensor The signal reception time difference and the position of the tool with respect to each AE sensor are used as input variables of the condition section, and this is used to infer breakage or abnormality of the drill by fuzzy inference. By installing the AE double signal AE sensor, it is possible to provide a highly reliable tool breakage detection device that can reliably detect tool breakage without being affected by the type of tool or cutting conditions due to differences in position.

[Brief explanation of drawings]

The figures relate to an embodiment of the present invention. Fig. 1 is a block diagram of a tool breakage detection device according to an embodiment of the present invention, and Fig. 2 shows the AE multiplied signal amplitude during normal cutting with a drill and due to the breakage. Figure 3 is a diagram showing the frequency bands of the power output due to drill breakage and the power output during normal cutting, and Figures 4(a) to 4(e) are the fuzzy waveform diagrams shown in Figure 1. FIG. 5 is a diagram showing the membership function of the variables of the condition part stored in the control part, and FIG. 5 is a diagram showing the membership function of the variables of the conclusion part stored in the fuzzy control part of FIG. 2...Workpiece, 4...Drill, s, io...A
E sensor, 28... fuzzy control section.

Claims (1)

[Claims] (1) A means for generating a pseudo AE signal containing the frequency component of the AE signal generated when a tool such as a drill breaks, and a means for detecting the AE signal provided near the workpiece in the main body of the machine tool for machining the workpiece. at least two AE sensors, and based on the detection signals from both AE sensors, the frequency component of the AE signal, the rise characteristic, the reception time difference of the AE signal at each AE sensor, and the distance from the tool to each AE sensor. A tool breakage detection device, comprising circuit means for data processing each distance difference, and wherein the circuit means is equipped with a fuzzy control section that fuzzy infers breakage of the tool using the data output of the circuit means.