JP2008087092A - Abnormality detecting device for tool - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an abnormality detecting device for a tool, which is easily applied to an existing tool, and certainly detects even a slight abnormality of the tool without erroneous detection. <P>SOLUTION: A machine toll cuts a workpiece W by the tool T with a single blade. The tool T is driven by a driving device M. A driving control portion 2 controls the driving device M so as to forwardly move from a fixed position toward the workpiece W, and separate from the workpiece W and backwardly move to return to the fixed position after cutting machining of the workpiece W. A contact detecting portion 3 detects whether the tool T contacts with the workpiece W from the output waveform of a vibration sensor 1 detecting the vibrations of the tool T. A cutting time measuring portion 4 detects cutting time during which the tool T contacts with the workpiece W, and a determining portion 5 determines that an abnormality is generated in the tool T when the cutting time deviates from a normal range. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a tool abnormality detection device that detects tool chipping in a machine tool such as a lathe or a thread cutting machine that performs cutting of a workpiece using a single-blade tool.

2. Description of the Related Art Conventionally, in machine tools that perform cutting, an abnormality may occur during machining due to chipping of a tool blade. As a technique for detecting this type of abnormality, a technique for monitoring the load current of a motor that is a drive source of a machine tool (see, for example, Patent Document 1), a sensor for detecting acoustic emission in the vicinity of a tool or a workpiece, A technique for monitoring the level of a specific frequency in the sensor output (for example, see Patent Document 2) is considered.
JP 60-56853 A Japanese Patent Laid-Open No. 1-140058

  By the way, in the technique described in Patent Document 1, since the load of the motor is monitored, there is a problem that the load increases / decreases due to not only the chipping of the tool but also the generation of chips, and erroneous detection increases. . In the technique described in Patent Document 2, since the level of a specific frequency generated as acoustic emission is detected, the vibration component of the machine tool may be mistaken as acoustic emission. In order to solve this type of problem, Patent Document 2 suppresses the transmission of bearing noise to the sensor by using a magnetic fluid and a magnet in order to reduce bearing noise.

  The present invention has been made in view of the above-mentioned reasons, and its purpose is to easily cope with an existing tool without adopting a special structure, and to reliably detect even a slight abnormality of the tool without erroneous detection. It is an object of the present invention to provide a tool abnormality detection device that can be detected at the same time.

  The invention of claim 1 is an apparatus for detecting an abnormality of a tool in a machine tool that approaches a workpiece from a position away from the workpiece and cuts the workpiece with a single-blade tool and then leaves the workpiece. A drive control unit that controls the drive device so that the drive time until the tool is moved backward from the workpiece to the fixed position after moving forward from the workpiece to the fixed position, and a vibration sensor that detects the vibration of the tool, A contact detection unit that detects a state in which the tool is in contact with the workpiece from the output waveform of the vibration sensor, a cutting time measurement unit that measures a cutting time in which the tool is in contact with the workpiece by the output of the contact detection unit, and a cutting time A cutting unit measured by the measuring unit is compared with a predetermined normal range, and a determination unit that determines an abnormality when the cutting time deviates from the normal range is provided.

  According to a second aspect of the present invention, in the first aspect of the invention, the determination unit sets a normal range of cutting time based on a distribution of cutting time measured by the cutting time measuring unit using a normal tool. It is characterized by that.

  According to the configuration of the first aspect of the invention, the cutting time in which the tool is in contact with the workpiece is detected by the output of the vibration sensor, and the cutting time is compared with the normal range. It is possible to detect the presence or absence of tool abnormalities simply by attaching the. Machine tools use single-blade tools and include lathes, boring machines, threading machines and the like. In this type of machine tool, when the tool feed is automatically controlled at a predetermined time and speed as with an NC lathe, if a normal tool is used, the cutting time for the tool to cut the workpiece is almost constant. The normal range of cutting time can be set. On the other hand, if the tip of the tool (bite) is broken or the tool is worn, the cutting time will be shorter than the normal range, so if the cutting time deviates from the normal range, It can be determined that an abnormality such as a broken blade or wear has occurred. That is, it is possible to detect the abnormality of the tool with higher accuracy than when detecting the load current of the motor.

  According to the configuration of the invention of claim 2, since the normal range of the cutting time is determined using the distribution of the cutting time when the workpiece is cut using a normal tool, the normal range is objectively determined. Therefore, it is possible to determine whether there is an abnormality in the tool based on a certain standard with no variation in the determination standard.

  A machine tool according to an embodiment described below is a cutting machine that includes a single-blade tool and automatically advances and retracts the tool with respect to a workpiece according to a program. An NC as an example of this type of cutting machine A lathe is assumed. Therefore, the tool is a bite, and the tool is advanced and retracted by the driving device separately from the driving source for rotating the workpiece.

  The tool abnormality detection device described in the present embodiment includes a vibration sensor 1 that detects vibration of the tool T as shown in FIG. If the vibration sensor 1 is directly attached to the tool T, vibration of the tool T can be detected with high sensitivity. However, since the vibration sensor 1 needs to be attached and detached when the tool T is worn or broken, the vibration sensor 1 Is attached to a holding base H for holding the tool T. The holding table H is moved by the driving device M to move the tool T forward and backward relative to the workpiece W. Further, the drive device M has not only a function of moving the tool T forward and backward relative to the workpiece W but also a function of feeding the tool T along the cutting surface.

  The position at which the tool T stands by without cutting the workpiece W is set to a fixed position. In the cutting process, the tool T is advanced from the fixed position toward the workpiece W, and the tool T is brought into contact with the workpiece W. Then, the workpiece W is machined, and after the machining is completed, the tool T is detached from the workpiece W and the workpiece W is retracted to a fixed position. The tool T is fed along the cutting surface during the processing of the workpiece W. In addition, according to an instruction from the drive control unit 2 that controls the drive device M, the drive time from the time when the tool T starts to advance from a fixed position, moves backward after machining, and returns to the fixed position is set to a fixed time by the program. .

  Therefore, as shown in FIG. 2, the movement time T1 while the tool T advances from the fixed position and contacts the workpiece W, and the movement time T2 while the tool T leaves the workpiece W and moves back to the fixed position. The total drive time with the cutting time T3 when the tool T is cutting the workpiece W is a fixed time. The cutting time T3 includes a time for feeding the tool T along the cutting surface.

  In the vibration sensor 1 described above, vibrations different from the movement times T1 and T2 are detected at the cutting time T3 when the tool T is in contact with the workpiece W. Therefore, if this state is detected, the tool T is applied to the workpiece W. It can be determined whether or not they are in contact. Therefore, the output of the vibration sensor 1 is given to the contact detection unit 3, and the contact detection unit 3 detects the state in which the tool T is in contact with the workpiece W based on the output waveform of the vibration sensor 1. A technique for determining whether or not the tool T is in contact with the workpiece W from the output waveform of the vibration sensor 1 will be described later.

  The determination result by the contact detection unit 3 is given to the cutting time measuring unit 4, and when the state in which the tool T is in contact with the workpiece W is detected, the timing of the cutting time T3 is started, and when the tool T is detached from the workpiece W. Time measurement of the cutting time T3 ends. Therefore, the cutting time T3 can be measured by the cutting time measuring unit 4. As described above, since the drive time (T1 + T2 + T3) from when the tool T moves forward from the fixed position toward the workpiece W, back from the workpiece W and returns to the fixed position is a fixed time, it moves when the cutting time T3 changes. Time (T1 + T2) changes. Such a change in the movement time (T1 + T2) also occurs when the tip of the tool T is chipped or the tip of the tool T is worn, so the cutting time T3 changes beyond the allowable range. If so, it can be determined that an abnormality has occurred in the tool T.

  Therefore, a normal range is determined in advance for the cutting time T3 and set in the determination unit 5, and the determination unit 5 determines whether or not the cutting time T3 measured by the cutting time measurement unit 4 is within the normal range. When the cutting time T3 is out of the normal range, it can be determined that some abnormality has occurred in the tool T. In particular, when the tool T has a chip or wear, it is considered that the moving time (T1 + T2) of the tool T becomes longer and the cutting time T3 becomes shorter as shown in FIG. When the cutting time T3 is shorter than that, it can be determined that the possibility of the chipping or wear of the tool T is high.

  Here, the normal range may be determined based on the distribution of the cutting time T3 when the workpiece W is cut using the normal tool T. For example, by applying this distribution to a normal distribution, it is possible to formulate such that the range of dispersion is set as a normal range, or a range three times the variance is set as a normal range. If the normal range is determined by this method, the normal range of the cutting time T3 can be objectively set.

  By the way, in order to determine whether or not the tool T is in contact with the workpiece W, it is necessary to detect a feature that appears along with the contact with the workpiece W in the output of the vibration sensor 1. The simplest technique is to detect the output amplitude of the vibration sensor 1 and determine that the tool T has contacted the workpiece W when the output amplitude exceeds a specified threshold value. However, when this technique is adopted, a criterion for how to set the threshold is necessary.

  On the other hand, whether the tool T is in contact with the workpiece W by extracting power for a plurality of frequencies from the output of the vibration sensor 1 (that is, extracting frequency components) and classifying the feature amounts using the frequency components as feature amounts. If a technique for determining whether or not is used, an accurate determination can be made. Therefore, the contact detection unit 3 uses an unsupervised competitive learning type neural network (hereinafter, simply referred to as “neural network” unless otherwise required), and gives the above-described feature amount to the neural network. It is desirable to employ a technique for detecting whether or not the workpiece W is in contact.

  As a neural network, it is possible to use a back-propagation type with supervision, but a competitive learning type neural network is simpler in configuration than a back-propagation type neural network, and learning data for each category is used. It is only necessary to learn by using, and it is possible to reinforce learning by additional learning even after learning once. The neural network is assumed to be realized by executing an appropriate application program on a sequential processing type computer, but a dedicated neurocomputer can also be used.

  The operation of the neural network has a learning mode and an inspection mode. After learning using appropriate learning data in the learning mode, the feature amount of the target signal to be classified in the inspection mode is given. Can be classified. That is, the feature amount obtained when the workpiece W is cut using a normal tool T is used as learning data, and the neural network is learned using this learning data. After that, the neural network is set to the inspection mode, and the characteristic amount obtained from the output of the vibration sensor 1 is sequentially given to detect the time point when the tool T contacts the work W and the time point when the tool T leaves the work W. it can.

It is a schematic structure figure showing an embodiment of the present invention. It is a figure which shows the operation principle same as the above.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Vibration sensor 2 Drive control part 3 Contact detection part 4 Cutting time measurement part 5 Judgment part H Holding stand M Drive device T Tool W Workpiece

Claims (2)

  A device that detects tool abnormalities in a machine tool that approaches a workpiece from a position away from the workpiece, cuts the workpiece with a single-blade tool, and then moves away from the workpiece, and advances the tool from a fixed position toward the workpiece. A drive control unit that controls the drive device so that the drive time until the tool is retracted to a fixed position after machining the workpiece becomes a fixed time, a vibration sensor that detects the vibration of the tool, and the output waveform of the vibration sensor. A contact detection unit that detects a state in which the tool is in contact with the workpiece, a cutting time measurement unit that measures a cutting time in which the tool is in contact with the workpiece by an output of the contact detection unit, and a cutting measured by the cutting time measurement unit A tool abnormality detection device comprising: a determination unit that compares a time with a predetermined normal range and determines an abnormality when the cutting time deviates from the normal range   2. The tool abnormality detection according to claim 1, wherein a normal range of the cutting time is set based on a distribution of the cutting time measured by the cutting time measuring unit using a normal tool. apparatus.

Images